0% found this document useful (0 votes)
34 views

Israels Edge

The story of the IDF’s Most Elite Unit - Talpiot.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
34 views

Israels Edge

The story of the IDF’s Most Elite Unit - Talpiot.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 228

Thank You for Downloading this Gefen Publishing House Ebook.

Sign up for our free weekly newsle er and get special offers on
exci ng new tles about Israel, Judaism and the Jewish People.

To sign up click here!


Copyright © Jason Gewirtz
Jerusalem 2016/5776

All rights reserved. No part of this publica on may be translated, reproduced, stored in a retrieval
system or transmi ed, in any form or by any means, electronic, mechanical, photocopying,
recording or otherwise, without express wri en permission from the publishers.

Cover Design: Joe Po er


Typese ng: Irit Nachum

ISBN: 978-965-229-713-6

135798642

Gefen Publishing House Ltd.


6 Hatzvi Street
Jerusalem 94386, Israel
972-2-538-0247
[email protected]

Gefen Books
11 Edison Place
Springfield, NJ 07081
516-593-1234
[email protected]

www.gefenpublishing.com

Printed in Israel Send for our free catalog

Library of Congress Cataloging-in-Publica on Data

Names: Gewirtz, Jason, author.


Title: Israel’s edge : Talpiot, the IDF’s most elite unit / Jason Gewirtz.
Descrip on: Jerusalem : Gefen Publishing House Ltd., [2015]
Iden fiers: LCCN 2015017950 | ISBN 9789652297136
Subjects: LCSH: Tokhnit Talpiyot (Israel) | Israel. Tseva haganah le-Yi?sra?el—Officers—Training of.
Classifica on: LCC U660.I75 G49 2015 | DDC 355.3/43—dc23 LC record available at
h p://lccn.loc.gov/2015017950
This book is dedicated to:
My father, the most honorable person I’ve ever known
My mother, the most selfless person I’ve ever known

My brother, the strongest person I’ve ever known


My wife, the most caring and wonderful person I’ve ever known

My two girls…be whoever you want to be, but be good and do good
To all of Talpiot’s past and future graduates, and to all the men and women who have served in
the Israeli army; you not only guard a na on’s border, you protect a people.
INTRODUCTION

Each year fi y thousand Israelis reach dra age. And each year staffers in
the Israel Defense Forces’ manpower offices start poin ng young men and
women in different direc ons.
To tank units. To the infantry. To the air force. To the navy. To ar llery
divisions. To special units designed to educate the underprivileged. To
pioneering units, engineering, intelligence; the list goes on and on. Men
are assigned to units for at least the next three years; women, at least the
next twenty months.
This is a country that needs a strong figh ng force perhaps more than
any na on in the world, and the army is idolized by its youth. In the years
before their induc on, mo vated Israeli teenagers jockey for posi on,
boos ng their high school grades and preparing for special IDF induc on
tests. Hoping to earn invita ons to enlist in the units they’ve set their
sights on, thousands of young Israelis even sign up for physical fitness
coaching and private pre-army training in order to prepare themselves for
the rigorous physical IDF exams that will decide who is sent where.
It used to be the air force that had first pick of these capable young men
and women. They needed the brightest and most physically fit to fly
complicated fighter jets, large transport planes and heavily armed silent
a ack helicopters – the figh ng force that makes up the long arm of the
Israeli Air Force. It was rare that anyone would say no to the air force. Like
American boys who dreamed of becoming a baseball star, many Israeli
boys hoped to grow up and serve their country by pilo ng an F16.
If the air force wasn’t your goal, or if your eyesight wasn’t perfect,
another highly desired spot was a place in Israel’s celebrated elite
commando units such as Sayeret Matkal or the paratroopers, with their
dis nc ve red boots that symbolize to everyone just how tough you are
and how dedicated you are to your country.
But in 1979 something began to change. While the air force and top
ground units were s ll, and always will be, a very high priority for Israel,
another unit – a secret unit – took over as top priority for the IDF.
Soldiers picked for this unit aren’t expected to fight like ordinary
soldiers. Instead, they are expected to learn. They are selected for an ultra-
compe ve, grueling, fast-paced program designed to give them the best
educa onal and military training possible.
Their teachers are some of the world’s top minds in mathema cs,
physics and computer science, as well as some of Israel’s top strategic
leaders from the army, navy and air force. The end game is to produce
soldiers with great military and scien fic backgrounds who are trained to
think like nobody else in the world.
They do not enlist like every other soldier in the IDF for three years or
twenty months. They sign up for a full decade. Because the program is so
intense, almost one in four of these bright minds originally selected won’t
make it through the program.
Graduates of the unit are expected to accumulate massive amounts of
knowledge and apply it directly to help Israel’s deterrence shield, to help
Israel’s intelligence services and to make the army, navy and air force
stronger in any conflict.
This small, select group is expected to change the way Israel does ba le.
The country is coun ng on them to give the Israel Defense Forces an
eternal edge over its enemies by developing the weapons and military
hardware of the future.
Since this unit became a part of the IDF, no other group of soldiers has
had such a profound impact on Israel – on Israel’s defense doctrine, on
how Israel’s weapons are developed and how they’re used.
The military advantage these special graduates gave Israel didn’t end
when their army service was over. Those who le the army o en took
what they did for the IDF and applied it to the Israeli economy, crea ng
hundreds of billions of dollars of wealth and tens of thousands of jobs in
Israel and beyond. They have drama cally helped give Israel an edge on
the ba lefield and in global business.
This is their story, the story of Talpiot.
CONTENTS

Introduc on

Chapter 1: A Devasta ng Force for Change


Chapter 2: Talpiot’s Founding Fathers
Chapter 3: Finding the Super Soldier
Chapter 4: Polishing the Program
Chapter 5: It All Starts in High School
Chapter 6: The World’s Fastest Learning Curve
Chapter 7: Training to Think Far Beyond the Box
Chapter 8: Reality Check
Chapter 9: A ack by Keyboard
Chapter 10: Making an Impact
Chapter 11: High-Tech Tinkerers
Chapter 12: Talpiots in Space
Chapter 13: Missile Command
Chapter 14: On a Mission
Chapter 15: Israel’s New Heroes
Chapter 16: “Talpiots Only Need Apply”
Chapter 17: Project Success
Chapter 18: Life Savers
Chapter 19: Reunion!
Chapter 20: The Future

Appendix: Timeline
A Salute from the Author
Acknowledgments
CHAPTER 1
A DEVASTATING FORCE FOR CHANGE

O ctober 1973: “The source” tells his Mossad handler in London that war
between Israel and its Arab neighbors is imminent. He’s been wrong
before, but with all of the other signs on the Syrian and Egyp an sides of
the border, it looks like this could be the real thing.
Agents from Britain to Israel are buzzing as they work their way up the
chain to the director of the Mossad, Zvi Zamir. He flies to London
immediately to meet with “the source” – a man now believed to be Ashraf
Marwan, the son-in-law of former Egyp an president Gamal Abdel Nasser.
The Mossad and Israel’s other intelligence agencies had their doubts about
“the source,” suspec ng him of being a double agent. To this day the truth
has not been revealed.
Zamir believes the informa on is credible and hopes the wheels of
Israel’s war machine will begin to move forward. He is able to convince
Israel’s Prime Minister Golda Meir and Minister of Defense Moshe Dayan
of the impending a ack.
Despite having that crucial informa on, Israel’s leadership decides not to
act – fearing it will be blamed for firing the first shot and that Israel will
thus lose cri cal support from the United States. Secretary of State Henry
Kissinger has warned Israel’s leaders – including Prime Minister Meir – that
if war should break out, Israel had best make sure it isn’t the one who
starts the figh ng.
At noon, October 5, Israeli military intelligence reports state, “The
probability that the Egyp ans intend to renew figh ng is low. There is no
change in our es mate of Syrian inten ons.”
October 6, 1973 – Yom Kippur, the tenth of the Hebrew month of Tishrei,
5734. At 2:00 pm, Egypt a acks Israel. Syria follows suit a few minutes
later.
Israel is caught off guard as Egypt begins its assault, crossing the Suez
Canal, which had separated Egyp an and Israeli forces since 1967. The
pace of Egypt’s advance across the canal picks up speed amid cries
broadcast from loudspeakers on Egypt’s side of the canal, “Allahu akbar!
Allahu akbar!” (God is great! God is great!). Israeli troops sta oned along
the Bar Lev Line – constructed as an “unbreakable defense” against Egypt
a er the war in 1967 – prepare to defend their lives and their country
against the onslaught. For many Israelis trapped on the Bar Lev Line, this
will be their final day.
In the north, Syrian tanks pour across the border that had marked the
cease-fire line established six years before. Syrian commando forces are
helicoptered to Mount Hermon, taken by Israel in 1967. Syrian fighters and
bombers fly sor e a er sor e, bombarding so and hard Israeli targets.
Confusion reigns from the northern p of Israel in the Golan Heights to
the Sinai desert in the south. Field commanders desperately try to stop the
simultaneous advances from the north and south.
Government leaders in Jerusalem are stunned into silence and worse,
inac on, while army leaders in Tel Aviv ini ally dismiss reports of the
a acks as exaggera ons. It is impossible to believe their Arab enemies are
capable of launching such swi and effec ve incursions. Didn’t these same
Israeli officers brilliantly thrash six Arab armies just a few years earlier?
Defense Minister Moshe Dayan, the highest ranking military official in the
country, leads his closest advisers to believe that there is no problem; the
situa on is under control and the Israel Defense Forces will quickly turn the
de.
Sirens blast throughout Israel on this, the holiest day on the Jewish
calendar. Nearly all Israelis, religious or not, mark Yom Kippur by forgoing
work, a ending synagogue services, fas ng, praying and contempla ng.
Bewildered Israelis stream out of their homes and synagogues. They gather
around radios wai ng for news, anxiously listening for the codes that will
instruct reserve units where to gather.
(Israel’s military brass had always contended it would need forty-eight
hours to mobilize reserves, the backbone of the Israeli army. And
intelligence agencies had always promised that giving forty-eight hours’
no ce would not be a problem. Dead wrong.)
Israeli reservists and volunteers commandeer cars and buses, hitching
rides. Ci zens beyond army age has ly arrange carpools to get soldiers to
the places where hundreds of units have been told to assemble for orders.
As the confusion in Israel grows, traffic backs up on the roads. Israel
approaches panic mode while Egypt and Syria approach the first-day goals
they set for themselves during the months of planning that led to “The
October War,” as it is referred to in Arabic.
Forty-eight hours into the figh ng, Israel is suffering unprecedented
losses on the ba lefield and the poli cal leadership is having a hard me
coming to grips with the turn of events. Sobering intelligence reports reach
Dayan detailing heavy losses and reports of Israeli posi ons overrun.
Dayan’s confidence evaporates.
Egypt now has hundreds of tanks and thousands of troops on the Israeli
side of the Suez Canal – an unthinkable scenario just a few days earlier. In
the north, the Syrian advance con nues. Syria’s tanks close in on the edge
of a plain that stares down on Haifa and the rest of the Israeli coast.
As Dayan finally realizes the level of danger to his country, he gives
chilling assessments of the situa on to his inner circle and to other high-
ranking members of the government. In his epic book The Yom Kippur War,
Abraham Rabinovich writes about an off-the-record mee ng between
Dayan and the editors of Israel’s newspapers in an a empt to level with
the na on. Sources told Rabinovich that Dayan said, “Israel might be
forced to withdraw deep into the Sinai.… The world has seen that we are
not stronger than the Egyp ans. The aura, and the poli cal and military
advantage of it being known that Israel is stronger than the Arabs and that
it would beat them if they go to war, this has not been proven here” (page
270).
A shaken Dayan is set to deliver a similar message to the na on that
night. But one of the newspaper editors fears that the general will further
erode morale and alarm everyone in the country. He reaches Prime
Minister Meir and advises her to have another high-ranking military officer
make the address to the na on.

***
How did this fiasco happen? When the Egyp an surprise a ack began on
October 6, there were only about a dozen Israeli tanks protec ng the road
that led from the Sinai straight into Tel Aviv. How did Israel – mighty Israel,
which had so thoroughly routed its enemies in the Six-Day War of 1967 –
fall to such a low state of preparedness only six years later?
In the years between the Six-Day War and the start of the Yom Kippur
War, Israel was engaged in a constant war of a ri on with Egypt and Syria.
There was daily shelling from behind the Arab lines targe ng Israel’s
military on the 1967 border and residen al communi es not far from
those lines, especially in the north of Israel.
Many na ons that previously had supplied arms to Israel cut off arms
shipments due to Arab threats. This was especially true of France, Israel’s
main weapons supplier before 1967. The French were told that if they
supplied Israel with weapons, the Arabs would cut off oil shipments to
France. The United States picked up some of the slack, but not all of it, and
Israel was le without a major weapons supplier.
At the same me, Arab na ons – especially Egypt and Syria – were
showered with arms by the Soviet Union, as the USSR tried to strengthen
its hold in this complicated, energy rich, and cri cal part of the world. In
addi on to its immense success in resupplying their armies, Egypt and
Syria made strides in training crews to man new weapons systems and
rapidly advanced in planning strategy.
All this happened amid an air of complacency in Israel. The country was
s ll res ng and sa sfied with itself a er the massive victory in 1967;
ci zens, army officers and the government thought there was no way Israel
could be defeated.
That smugness came to a swi end when war broke out in 1973. In the
opening days of figh ng, Israel lost forty-nine planes (compared to forty-six
planes during the en re Six-Day War). By the me the Yom Kippur war
ended in late October, Israel had lost almost a fi h of its air force. Almost
all of those planes lost by the Israeli Air Force were shot down by Soviet-
made surface-to-air missiles. The Russian SAMs were so proficient that
Israeli pilots would call them “flying telephone poles,” alluding to the wires
that could snag a low-flying fighter jet – wires that were impossible to
evade at a low al tude.
Those Soviet SAMs were so new and sophis cated, no air force in the
world could dodge them. The Israeli military came to the shocking
realiza on that the IAF was no longer in complete control of the skies.
Israel’s defense doctrine had been built on tanks and airplanes, and both
were being soundly defeated by new technology owned by the Arabs.
Air defense wasn’t the only area where Arab na ons had advanced their
figh ng technologies. Arab foot soldiers defeated Israeli tank columns with
brand new, portable, AT-3 Sagger wire-guided an -tank missiles. (The
Russians call them 9K11 Malyutka missiles.) They were carried into the
Sinai desert by Egyp an troops in suitcase-sized containers. Before the
shoo ng started on October 6, Egyp an special forces passed through the
Israeli frontline on the Suez Canal with their new and deadly cargo.
Israel’s high command had some knowledge of the Sagger, learned from
American figh ng experiences in Vietnam. But high-ranking Israeli officers
did not believe Arab fighters would be strong enough or brave enough to
confront Israeli tanks. They knew that in one-on-one ba le an Israeli tank
crew could usually defeat its Egyp an or Syrian counterpart. Israel’s tank
crews were generally be er educated and be er trained than Arab crews;
moreover Israel’s tanks were more accurate and could fire from a longer
distance.
But they had not considered the Egyp an army’s ability to use the
sophis cated Saggers, and Israeli tank crews were caught totally
unprepared. A er stealthily crossing the canal, the Egyp ans set up shop,
hiding behind sand dunes and se ng up traps that would easily ensnare
formidable Israeli tank units.
Military experts later confirmed that “the Sagger operator would find it
much easier to hit a tank than the other way around, and at ranges that
matched the tanks” (Rabinovich, page 36). The large Israeli tanks made
invi ng targets for Sagger marksmen: Egyp an specialists would lock onto
Israeli armor more than a mile away, fire and destroy tank a er tank with
deadly accuracy. In contrast, Israelis searching for Sagger crews in the vast
desert sand could rarely locate them. This new element to the ground war
would cost Israel dozens of tanks in the opening days of figh ng.
In the crucial early days of figh ng, it became clear that counter a ack
and reinforcement of Israeli troops would be impossible. Israeli tank
reinforcements posi oned ten miles away from the canal were easy prey
for those hidden Sagger crews. When reserve tank forces tried to push
forward to Israeli posi ons along the canal, they were picked off one by
one.
Communica ons failed as well. Less than forty-eight hours a er the war
began, with eyewitnesses repor ng Syria and Egypt were using machine
guns to kill scores of Israelis captured in the opening hours of the war,
Defense Minister Dayan found himself having trouble ge ng through to
commanders on the ground.
At the same me, the air force was planning an opera on to try to
destroy Egypt’s Soviet made an -aircra ba eries. Minutes before the
mission was set to launch, Dayan contacted Benny Peled, commander of
the Israeli Air Force, and called it off, diver ng all planes to the north to
stop the Syrian advance. His logic was that there was only sand in the Sinai
between Israel proper and Egyp an tanks in the south. In the north, Israeli
civilians were about to be in Syrian range: without the air force, those
popula on centers, including Haifa, were doomed.
During the 5:00 am conversa on with Peled on October 7, Dayan said,
“If our planes are not a acking by noon the Syrians will reach the Jordan
Valley.” Then for the first me Dayan used a phrase that he would repeat in
the coming days to the dismay of all who heard him. “The Third Temple,”
he told Peled, “is in danger” (Rabinovich, page 175).
“The Third Temple” was, and is s ll, code to many represen ng modern-
day Israel. The first two Holy Temples in Jerusalem were destroyed in
ancient mes; the first by the Babylonians in the year 586 BCE, the second
by the Romans in the year 70 CE. To many secular Israelis, as well as Jews
all over the world, today’s Israel is the “Third Temple.”
The intelligence failures that led to Israel’s early problems in the Yom
Kippur War were drama c and are s ll referred to today, more than forty
years a er the war, but the war wasn’t a total surprise to everyone in
Israel. Some intelligence officers had seen signs that war was imminent.
Soviet advisers and diplomats had moved their families out of the country
in the days before October 6. There were widespread troop mobiliza ons
in both Egypt and Syria; Arab troops were on the move, en masse.
There was also somewhat confusing evidence that the Egyp an and
Syrian units mobilizing near the front lines were ge ng instruc ons to
carry out “training” exercises. The idea was to move troops into posi on,
but allow Israeli spies and communica on eavesdroppers to believe it was
for the purpose of training only – not war. However, a more careful Israeli
inspec on would have allowed the Israelis to see that those troops weren’t
actually carrying out exercises at all: they were simply preparing for an
invasion. If all this weren’t enough, there was a direct warning from
Jordan’s King Hussein to Prime Minister Meir in person during a secret
face-to-face mee ng.
In modern warfare, the side that fires first o en has the upper hand.
Wai ng to be a acked is especially risky for a ny country like Israel, which
is only sixty miles wide at its broadest point. At its narrowest point, Israel is
roughly ten miles wide. Even a moderate thrust by the enemy could cut the
country in half in the first hours of any war.
Of course, Prime Minister Meir knew all of this, but decided to wait and
hope for the best rather than risk diploma c blowback, especially from the
United States. She had been assured over and over by her military advisors
– including revered Defense Minister Dayan – that if war were to break out,
Israel would be able to smash its enemies again as the country had done in
1967.
When the shoo ng stopped and cease-fire agreements were signed
three weeks later, Israeli ci zens, the government and army woke up to a
bi er reality. They were not invincible.
Two thousand six hundred fi y-six Israeli soldiers had been killed.
Almost nine thousand others lay wounded. The backbone of the IDF – its
formidable tank force – had lost two hundred of its three hundred tanks in
the Sinai in the first twenty-four hours of the war. Scores of Israel’s fighter
planes were gone.
The Arab armies had not only caught up, but had surpassed Israel in
technology, subterfuge, strategy and opera onal capability. Israel also lost
in an area that was unforgivable: they lost the intelligence war.
The trauma of the Yom Kippur War and the realis c fear that another
war could mean ex nc on lingered. In the war’s a ermath, the en re
military establishment and poli cal leadership were forced to resign. The
prime minister and Defense Minister Dayan both le eight months a er
the war ended. Key players in intelligence and in the IDF were ousted. For a
na on born out of the ashes of the Holocaust to be threatened in such a
way has a deep, scarring and las ng impact that affects every aspect of
public life. A country of this size, surrounded by so many enemies with
much larger armies and with suppliers like the Soviet Union, simply could
not risk another war such as this. For years, Israel existed in a state of quiet
tension and nobody really knew how to return to the sense of security
most of the popula on had felt just a er the Six-Day War.
Everyone knew that Israel would never have the numbers compared to
Egypt, Syria, Iraq, Jordan, Lebanon, Saudi Arabia, Iran, Algeria, Libya and
Sudan. As the deck forever will be stacked against Israel quan ta vely, it
became clearer than ever that Israel needed a qualita ve advantage.
It was not long a er the devasta ng Yom Kippur War that two professors
at Hebrew University had an idea to give Israel that qualita ve edge it so
desperately needed (and s ll needs today) to survive. Their goal was to
rearm Israel with enterprising minds, a weapon no army could defeat or
suppress. Those minds would supply Israel with advanced weapons ahead
of anyone in the world.
But the professors’ idea went beyond weapons. The concept was also to
train those young minds to come up with new and be er ways to monitor
the enemy and to outsmart it.
The inspired proposal wouldn’t just help for fy Israel for the next war.
Their innova on gave Israel the lead it maintains today, four decades a er
the Yom Kippur War of 1973, the edge Israel will have over its enemies for
decades to come.
CHAPTER 2
TALPIOT’S FOUNDING FATHERS

W hen the air-raid sirens began on October 6, 1973, Hebrew University


professor of physics Shaul Yatziv knew the government would never call for
a mock air raid on Yom Kippur. Something awful and unthinkable was
happening.
Most Israelis had heard word of threats coming from Egypt and Syria,
but like the government and the army, the civilians in the ci es, villages
and kibbutzim throughout the country knew li le about Arab inten ons.
But when the silence of Yom Kippur was broken by siren blasts, and radio
announcers calmly but urgently began reading the codes to mobilize
reservists, it could mean only one thing: war.
Armed conflict was not new to Yatziv. Born in pre-state Pales ne in 1927,
he had lived through war and the threat of sudden violence from the day
he was born. Following the declara on of Israel’s independence in May
1948, he served in the first Jewish army assembled in two millennia.
Yatziv was a full professor at Hebrew University, part of the university’s
Faculty of Natural Sciences; in 1973 he was conduc ng research on op cal
lasers and spectroscopy, the study of how ma er reacts to radiated energy
and vice versa.
Like everyone else in Israel, Yatziv was shocked by the Yom Kippur War
and its ramifica ons. At forty-six, he was too old to make a difference as a
soldier, but he had another idea, an idea that would give Israel a real edge
going forward. He knew that he would need to develop his concept and
collaborate with others to make it a reality. He thought of Professor Felix
Dothan, a colleague at Hebrew University with whom he had shared his
research over the past five years. Their focus on lasers and poten al future
applica ons had been promising.
But on Yom Kippur 1973, Professor Felix Dothan was far from the turmoil
in Israel. On loan to the University of California-Irvine from Hebrew
University, he was personally safe, but quite agitated. His homeland was
under a ack and he worried about his family and friends.
In most countries, residents under a ack run away from the figh ng. But
when war breaks out in Israel, Israelis living abroad o en run toward the
figh ng. They return from wherever they’re living, working or vaca oning.
When the war broke out in 1973, interna onal carriers cancelled all service
to the Middle East, but El Al flights were filled with Israelis rushing to get
home to help, to serve and to fight.
Professor Dothan was almost fi y years old in 1973. The country didn’t
need him in the army; it was more important for reservists of figh ng age
to have those airline seats. He would sit this war out, stranded in Southern
California, le to nervously follow developments by reading newspapers
and watching the evening news. He would have to wait pa ently for word
on the fate of his loved ones, including his son Yoav, who had joined the
army at the age of seventeen, just two months before the war broke out.
Usually soldiers aren’t eligible to join un l the age of eighteen, but because
of Yoav’s academic acumen and with his father’s wri en consent, he was
allowed to enlist early.
He was proud of his son’s patrio sm; it was the mark of a true Israeli.
For unlike Shaul Yatziv, Felix Dothan was not a na ve of Israel. He was born
Felix Deutsch in Zagreb, Yugoslavia, in 1924.
He had seen the dark side of life as a European Jew in his youth. Shortly
a er Yugoslavia had been occupied by the Germans, Felix and his
classmates were told that the next day they were to assemble in the forest,
rather than report to school. Felix’s father refused to give his permission
and Felix stayed home. They later learned that Nazis and their Yugoslavian
collaborators had shot four hundred of his classmates to death.
That wasn’t the only me Felix narrowly escaped death. Later, he was
arrested along with his family by the Nazis, but managed to avoid
deporta on to Auschwitz due to the family’s wealthy connec ons. Hidden
by gen les ll the end of the war, he finished high school and started at the
University of Zagreb, studying electrical engineering.
But a resurgent wave of an -Semi sm in Yugoslavia – and the realis c
fear of being stuck behind the Soviet Union’s Iron Curtain forever –
propelled his family to move to Pales ne while they s ll could. They
arrived in 1948.
The War of Independence against seven invading Arab na ons was
raging. The profound impact of his narrow escapes from the hands of an -
Semites evoked his deep loyalty to the newly emerging Jewish na on.
Deutsch joined the army and worked his way up the ladder to lieutenant
colonel in the Israel Defense Forces. A er the cease-fire in 1949, he
scratched out a living as a fisherman while con nuing his studies in
engineering and physics at the Technion in Haifa, gradua ng in 1951. (The
Technion is Israel’s version of the Massachuse s Ins tute of Technology in
the United States.)
He immediately took a posi on at the precursor to Israel’s now-famous
Rafael Advanced Defense Systems in Jerusalem. The exact nature of his job
remains classified, but it involved advancing the produc on, tes ng and
manufacturing of new weapons for the army.
He carried that experience to a new posi on, teaching and researching
at Hebrew University in Jerusalem. Hebrew U. is an auspicious place for a
researcher; one of its founders was the celebrated physicist Albert Einstein,
and it had the backing of philosopher Mar n Buber, psychoanalyst
Sigmund Freud and Israel’s first president, Chaim Weizmann. Along with
the Technion, it became the backbone of educa on in Pales ne soon a er
opening its doors in 1925.
Enjoying a gra fying career, he con nued his research in Switzerland,
later returning with his young family to Jerusalem and Hebrew University’s
new physics lab. There he worked with electrified gas and plasma, which
has many civilian and military applica ons. A er comple ng his doctorate
in 1965, he hit the road again, accep ng a posi on as a visi ng professor in
Geneva’s Ins tute for European Nuclear Research. There he specialized in
working on calcula ons regarding magne c fields, design and lasers.
A year a er the pivotal Six-Day War of 1967, he became a senior lecturer
at Hebrew University’s Racah Ins tute of Physics and changed his name
from the European-sounding Deutsch to a Hebraic version, Dothan.
A er fulfilling his role as visi ng professor at the University of California-
Irvine in 1973, Professor Dothan made his way back home to Hebrew
University. When he arrived, he was shocked to find the confident, happy
country he had le in 1968 in the midst of a crisis. Israelis had lost trust in
their government and in the military. The mood of his once-op mis c
na on had turned decidedly pessimis c.
Professor Dothan wanted to help get his beloved Israel back on its feet;
the ques on was how could he give such a gi to his na on. He knew the
answer had to do with research and educa on, not military muscle, but
even if he could come up with a viable plan, how could he get a military
machine that’s used to figh ng wars with tanks and manpower to listen?
It was me to revive collabora on with Professor Yatziv. Together, they
went to work, star ng with a posi on paper to submit to Israel’s top
military officers. The paper states: “Concern for the fate of Israel and the
wish to do the utmost to lower the number of casual es in future wars
mo vated us to submit a proposal that includes three important star ng
points that we do not have in exis ng research ins tu ons.”
Their star ng point: “Afirm recogni on that the State of Israel must
make an effort to develop totally innova ve weapons that do not exist
among the na ons. This goal can only be obtained through human
crea vity, crea vity that reaches its peak at the low biological age of the
twen es. Inven ve capability requires crea ve imagina on, vast
knowledge and deep understanding, but can be significantly encouraged
by presen ng challenges and crea ng a lively and mo va ng atmosphere
where every effort and contribu on will get encouragement from the
surrounding environment. As one of the ways to reach this goal we
propose a systema c and condensed effort to invent and develop new and
efficient weapons, where new is defined as a weapon that is not in use in
other armies, even in the armies of the superpowers. It is essen al that the
core of this program will be based on the most talented and dedicated
people, who also possess the required background in the natural sciences
and in the weapons technology.”1
Their second point was to propose that the Israel Defense Forces would
be directly responsible for the program and the people involved in it,
specifically the Israeli Air Force.
As men oned earlier, the air force was the most venerated branch of the
IDF. They took only the best of the recruits, people who could be entrusted
with the most expensive armaments in Israel’s arsenal. They needed math
skills, the ability to understand advanced physics and aeronau cs, great
test scores and the ability to think fast.
Part of Dothan and Yatziv’s proposal was to change that doctrine, to
push the brightest and most mo vated soldiers to their newly conceived
program.
In Israel most students don’t graduate university un l they are twenty-
five years old due to three years of post–high school compulsory military
service. If a student was accepted to Talpiot and then had to serve for eight
years (as their plan s pulated), he would be behind his peers for far too
long, pu ng him at a disadvantage throughout his career.
So point three of the original Talpiot proposal allowed for cadets to earn
a Bachelor of Science degree in physics and mathema cs (and later
computer science as well), disciplines needed for engineers working on
advanced weapons systems.
The original proposal called for cadets to earn those degrees in three
years rather than the four given to ordinary students studying those
subjects. A er comple ng their studies, they would serve in the army for
an addi onal five and a half years.
Ge ng started, however, wasn’t so easy. Dothan and Yatziv needed the
IDF brass on their side and not just the Ministry of Defense. They needed
support from army and air force generals, other high ranking officers in
different branches of the IDF and the all-important general staff.
Mee ng with top generals, even in a country as small as Israel, isn’t easy.
In the 1970s, the army was Israel’s most important ins tu on, and many
high-ranking officers didn’t really want advice from professors or civilians.
For three years, Dothan and Yatziv met with the military brass, only to
be denied at every turn. This was especially frustra ng to Dothan because
he had es to two of Israel’s top generals. Both General Haim Bar Lev and
General David Elazar originally came from Zagreb. Both had dis nguished
themselves in previous wars, and Elazar was the na on’s highest ranking
officer, the chief of staff, during the Yom Kippur War. But hometown
nostalgia was not enough to get Dothan through the door.
The truth was that in the wake of the Yom Kippur War, Elazar and all the
top brass had troubles of their own that preoccupied them. Inves ga ons
into military fumbling during the war were being conducted by a special
commission headed by Chief Jus ce of the Supreme Court Shimon Agranat.
The commission’s final report was issued in January 1975, and the
impact on Israel’s poli cal and military leadership was devasta ng. In
addi on to Golda Meir, many high-ranking officers – including Elazar –
were forced to leave the army, along with several of their top depu es.
New leadership was brought in, but it would take me to rebuild the
army’s infrastructure. While the leadership was changing, the IDF was
focused on quickly replacing armaments, tanks and fighter planes lost in
the war. In the confusion, it is likely that nobody in the army had the me
or the will to focus on a report from two outsiders.
In April 1977, with the country s ll reeling from the a er effects of the
Yom Kippur War, the Labor Party was swept out of office for the first me
since the founding of the modern state in 1948. Menachem Begin, of the
Likud Party, had led the opposi on for decades, and now it was his turn to
lead Israel.
Begin moved into the prime minister’s office ready to make changes.
Less than a year a er taking office, Begin’s defense minister, Ezer Weizman,
appointed Rafael Eitan as chief of staff, Israel’s highest ranking military
officer. General Eitan was one of the Israeli officers credited with stopping
the massive Syrian advance in the opening days of the Yom Kippur War. He
had lost many of his soldiers in the figh ng, but he escaped the war
unscathed professionally.
He had come from a disadvantaged economic background, but he
revered educa on and saw it as the key to improving the prospects of less
advantaged Israeli youth. With Eitan’s ascension to chief of staff, new life
was pumped into Dothan and Yatziv’s drive to form an elite educated unit
in the army.
Their idea worked its way up the new and improved chain of command
to the office of General Eitan. He was known to favor educa on, and
because the structure of the army was changing, he had extra incen ve to
try new programs that would bear fruit in the years, decades and
genera ons to come.
One day in 1978, a er years of trying to break through the military’s
roadblocks, Dothan and Yatziv were finally called to deliver their proposal
directly to Eitan. They arrived, proposal in hand, and spent a few minutes
with the diminu ve but imposing general. Then they were asked to wait in
an exterior office.
General Eitan had his secretary ask Air Force Colonel Benjamin Machnes
to come to his office immediately. A pilot in the early days of Israel, the
highly capable Machnes rose up the ranks. A er his flying days were over,
he headed a school that taught air force personnel advanced physics and
aeronau cs. He reported directly to the head of the Israeli Air Force,
General Benny Peled.
Eitan knew of Machnes’s work at the air force school. Given permission
by his commanding officer, Benny Peled, Machnes immediately reported to
General Eitan. He described their mee ng: “I opened the door and Raful
(nickname for Raphael Eitan) was wai ng with General Israel Tal, si ng
there. Tal was our top tank general; he invented the Merkava.… I said,
‘Hello, General Eitan, I’m Benji Machnes.’ Eitan replied, ‘Of course I know
you.’ He didn’t even ask me to sit. He said, ‘Outside my door are two
professors. I think they have a good idea. Go and do it. That’s all.’”
While Colonel Machnes was accep ng the posi on as the army’s first
representa ve in this new joint venture with Israeli academia, Yatziv and
Dothan waited anxiously. Machnes let himself out of the office and
introduced himself to the professors. He then told them, “Your project has
been accepted, let’s get to work.” Yatziv and Dothan looked incredulous.
“That’s it?” Machnes said yes, and the planning stages of Talpiot began
right there outside of the chief of staff’s office.
A short me later, General Ariel Sharon was named Minister of Defense.
He was quickly brought up to speed on the project. He told Machnes,
“Benji, this is a good thing that you’re doing.” By Machnes’s account,
neither Sharon nor Eitan were par cularly interested in becoming
in mately involved in the program, but both were willing to gamble on it.
Professors Dothan and Yatziv were given a mandate to start, and were
urged to launch their innova ve program in just a few months’ me. There
was so much that had never been discussed, so many ques ons looming
that would demand foresight, crea vity, prac cality and for tude. They
knew they would have to scramble to meet such a pressing deadline, but at
last they were underway.

Notes
1 IDF Archives.
CHAPTER 3
FINDING THE SUPER SOLDIER

Ttheheprogram.
first thing the professors and Colonel Machnes had to do was name
For years, Dothan had been thinking of an appropriate name,
and he suggested Talpiot. Talpiot has a few meanings in Hebrew, but its
most popular defini ons are “sturdy strongholds” or “tall turrets.” In the
biblical Song of Songs, talpiot appears as a metaphor for leadership.
(Though not religiously observant, many of Israel’s early leaders were
aware of the historic na onal significance of the Jewish homeland and took
pride in their new state’s connec on to the Bible. To this day, biblical
references are commonplace in Israeli public venues.) The others agreed to
the bold military tle.
In order to launch the program in a ma er of months, they had to
quickly come up with a viable program, a syllabus, a home, a university
partner, appropriate facili es, a way to find and recruit students, and a way
to test poten al applicants.
Dothan and Yatziv had both been insiders at Hebrew University for some
me. The army held mee ngs with several of the top universi es in Israel
including Hebrew University, Tel Aviv University and the Technion in Haifa.
There was outright rejec on, at first. None of the three universi es
wanted soldiers on their campus. An early leader of Talpiot started to think
the program would never get the backing of a university. One day, as he
was speaking to a friend who was a secretary at Hebrew University, she
asked him what was wrong. He told her the story and she exclaimed, “You
haven’t spoken to the right person!” Two minutes later, she came back
with a vice president of Hebrew University, Professor Yo’ash Vedyah, who
listened to the plan. He was so taken by it that he moved quickly to
convince the board to broaden its scope, allowing Hebrew University to
become the home of this new, mysterious and top secret army program.
This was not the first nor the last me that an Israeli secretary knew more
about making the right connec ons than the “experts.” It seems to be built
into Israeli culture.
Internal mee ngs following the decision at Hebrew University o en
descended to figh ng over the content of the program and whether it
would be academically rigorous enough. The university didn’t want to hand
out degrees to candidates they did not believe were worthy.
Several generals, including some in the general staff, wanted to kill
Talpiot in its infancy. They thought it would cost far too much money, and
that it was too eli st.
One of the IDF’s many redeeming quali es is that it is the great equalizer
in the na on. It didn’t ma er if you were smart or dumb, rich or poor, you
were in the na on’s army with people you might not ever normally come
into contact. The IDF truly is an army of the people, a place where
everyone is needed. The disgruntled generals felt that a program like
Talpiot would turn that line of thinking on its ear, and they were vocal in
their opposi on.
Colonel Machnes always defended the program, “I was always straight
to the point when people in the army said Talpiot wasn’t necessary. I would
always speak up, no ma er who was on the other side. [I would] say we
were important and the state needed Talpiot. I was constantly figh ng with
higher ranking officers in the opening years, officers who were trying to
squash our budget. There were numerous arguments.”
And to many generals, the thought of Talpiot becoming the army’s top
priority was abhorrent. They needed fighters. They needed mo vated
young Israelis to fly planes and drive tanks; they needed boots on the
ground, and they needed to secure the na on from the sea.
Talpiot’s recruiters were forced to jockey for posi on, smashing elbows
with other officers represen ng other elite branches of the army. There
was infigh ng. There was subterfuge. Bad and conflic ng advice was given
to poten al recruits.
Fortunately, the first commander of Talpiot was an indomitable
personality, Dan Sharon. His war experience had taught him that there was
a need for such a program well before the idea began taking off. On the
first day of figh ng in the Yom Kippur War, Egyp an warplanes fired
Russian-made Kelt missiles at his base. The Kelts were precursors to today’s
cruise missiles, boas ng pinpoint accuracy. Major Sharon saw some of the
reserve foot soldiers who were on his base come up with a quick way to
defeat and deceive the Kelts by tapping into their Israeli radar systems that
were on a similar frequency. “I thought, why are these guys here? They
should be at the Ministry of Defense. At this point, I knew we were was ng
resources and needed to do be er.”
Sharon was a long- me friend of Felix Dothan. (They used to meet on
Fridays for a spot of brandy.) He was very eager to help Dothan, so when
the program was finally accepted in 1978, Dothan asked him to lead Talpiot
and Sharon accepted. He had just completed his PhD disserta on at
Hebrew University on “the development of thinking and how one can
improve his or her own thinking.”
In order to take the Talpiot posi on, however, the army had to reac vate
him. They did so immediately, with the higher rank of lieutenant colonel
(sgan aluf, execu ve officer of a brigade).
Sharon looked to many places for help and advice; he’d never started a
military unit before. Benny Peled, former chief of the Israeli Air Force and
architect of the IAF’s stunning defeat of the Egyp an and Syrian air forces
in the Six-Day War, offered his services. Sharon recalls, “Peled was always
very posi ve, but cri cal and cau ous at the same me. He always had a
very sharp mind. He said to me, ‘Listen, when you construct a bridge, you
always leave a weak spot where you can destroy the bridge with one single
s ck of dynamite, if necessary. Maybe we are making a mistake, so don’t
forget: in case this program is unsuccessful, you need to get out.’”
Sharon quickly discovered there would be bumps on the road,
par cularly in finding the super soldiers he needed for Talpiot. He
discovered that he was in compe on with a preexis ng elite computer
and communica ons eavesdropping unit called Unit 8200. Its recruiters
were ac ve precisely where Talpiot hoped to a ract candidates.
“It became clear to us when we came to the field, the guys from 8200
were already there, and already had recruited.… An unpleasant situa on
developed and I decided to solve this problem. There was a colonel in
intelligence who was responsible for their training, Sasson Sahaik. At first,
he didn’t want to meet with me to discuss this issue, but ul mately he
agreed. We sat in the Harley Café in Tel Aviv. I looked him straight in the
eye, I told him this is about our existence here. Then I said, ‘Tell me – is it
good for us to fight? Let’s compromise. We’ll find them together and we’ll
ask each student what he wants. That’s it. If you have one or two who are
exactly for you, take them. But in the end, let the candidates do what they
want.’” Like many things in Israel, that informal deal was sealed with a
handshake, and the truce held.
In the early days, top generals also argued that army recruits shouldn’t
be studying in a classroom. But Colonel Machnes made sure they were
fighters first. He repeatedly pointed out to Talpiot detractors that “our
Talpiot recruits had Israeli Air Force uniforms and I insisted they’d be an
army unit, not students in the army. I ran it as a figh ng unit.”
Dothan, Yatziv and Colonel Machnes set the standards for admission and
recruitment very high. In one of their original organiza onal memos they
wrote:

We need applicants with a high IQ. We are looking for the top 5 percent
when it comes to intelligence, crea ve ability, the ability to focus, stable
and pleasant personali es; these people will need to be in constant
contact with the employees of Research and Development of the
Ministry of Defense, combat officers and professionals, scien sts in
ins tu ons of higher educa on and engineers and technicians in the
ins tute where they’ll work.… [Applicants must have] dedica on to their
homeland and the strong will to survive in the unit.2

That is clearly not an easy shopping list, especially when the new unit had
to compete with Special Forces units and the Israeli Air Force, which had
similar requirements. Both had two huge advantages over Talpiot. First off,
every poten al recruit had heard of both; and the smartest and most able
students had long strived to join them. Everyone knew that the training,
connec ons and pres ge of belonging to those units would later help them
in their post-army careers. Nobody had ever heard of Talpiot, not even
some of the army’s top officers, let alone the recruits.
Exis ng units had the added advantage of public image. Army service in
Israel was (and is s ll, though to a much lesser extent) seen as a coming of
age, when boys turn into men and girls become young women. Most
seventeen- and eighteen-year-olds don’t want to sign up to study. They
want to fight. They want to impress girls with their special insignias and
berets. They want to be seen as men protec ng their na on. To many,
si ng in a classroom is like si ng on the bench. They want to be in the
game.
Yet many of the top Israeli recruits would quickly find they didn’t have a
choice of where they’d go anyway. Chief of Staff Eitan wanted to make
Talpiot a priority, and he was going to get as many generals on board as
possible to make it happen – even if the recruits had never heard of
Talpiot.
A few years a er the program was created, it became common
knowledge among senior officers in the IDF that Talpiot would be given the
right of first refusal regarding all enlistees in the armed forces. If you were
accepted into the training program for air force pilots but Talpiot’s
commanders wanted you, you would go to Talpiot. You could be a fighter
pilot later, but first you were going into Talpiot.
The founders were working under the theory – backed up by research –
that they could only use young men (and later, women) of a young age
because they believed crea vity and the proclivity to believe “anything is
possible” peaks in the early twen es. If the young recruits wanted to do
something in the army in addi on to Talpiot, that wouldn’t be a problem.
But Talpiot must come first.
Recrui ng in those first years was rudimentary. Army human resources
officers would gather data about poten al candidates from other
recruitment officers through a large database. But because Talpiot didn’t
know exactly what criteria to judge, it was not an easy process.
Talpiot recruitment officers would also go to schools, mostly in Tel Aviv,
Jerusalem and Haifa, to talk to school principals and tell them a li le about
the program, hoping they could connect and find appropriate candidates
about to graduate high school and enlist in the army. But this was far from
a scien fic process, and many capable students and candidates living
outside of Israel’s three main ci es were le out of the mix. It took the IDF
years to figure out how to equalize the recrui ng process and cover the
na on’s smaller and less wealthy areas.
S ll, finding the right recruits was a problem. Dothan and Yatziv began
working on criteria to help them make sure the right candidates were
applying and being accepted. In the early years, they wanted to isolate
candidates who would be able to deal with a lot of new physics and
mathema cs material in a short amount of me, understand it to the
degree that they could apply it to actual projects, achieve a bachelor’s
degree and absorb the material usually given to excep onal students in a
four-year period – in just three years.
The original tests for Talpiot tested both cogni ve func on and
crea vity. (Later, a component designed to test for future success in a team
environment was added.) The entry tests were devised by experts in both
math and physics. In addi on, psychometric tests were devised to test for
intelligence, the ability to learn new subjects and, of course, personality
traits.
The students taking the tests in the early years of Talpiot recruitment
were o en puzzled. They had never been in this posi on before. These
were personal ques ons, not ques ons they had expected: How do you
feel about yourself? How do you feel when you don’t do as well as you had
hoped on an assignment? Do you feel nervous when addressing a group?
A er those first tests, the pool of several hundred applicants had to be
whi led down to several dozen. Now things really got interes ng. A few
months before the enlistment date, candidates were called in for personal
interviews. One by one, they were called into a room. The seventeen-year-
old high school student would sit before a panel of eight to ten people,
many of whom were high-ranking army officers, heads of the IDF from the
Defense Ministry’s Research and Development arm, MAFAT. (There will be
more at the end of this chapter on MAFAT, which helps oversee Talpiot in
conjunc on with the Israeli Air Force. More importantly, representa ves of
MAFAT are later very influen al in analyzing the Talpiot graduates and in
aiding in their choice of where they will serve in the army a er they
graduate from three years of intense coursework.)
The personal interview part of the tes ng phase usually lasted about
thirty minutes. The panel scru nized how the candidate acted under
pressure, how poised and crea ve he was, how he fielded ques ons and
his ability to communicate with older, more powerful and more
sophis cated men.
The interviewing commi ee became known as the Character
Acceptance Commi ee. There were many levels to “the interview,” but
staples included asking candidates ques ons regarding math and physics,
to find out how well the poten al candidates could understand new
material. In some cases, they were given material to read, then quizzed.
They were asked seemingly simple ques ons to find out how much they
enjoyed learning about science and how curious they were. Ques ons may
have included “How does an airplane fly?” and “How does a refrigerator
work?”
Haggai Scolnicov, a Talpiot graduate, reflected on the grueling test
phase, par cularly the commi ee interview. It seemed to him that they
were looking for leadership and a candidate’s ability to take on difficult
tasks, technological leadership in effect. For instance, “They ask you to
explain physical phenomena which are likely well beyond what anyone
ever studied in school and to do the best you can with a ques on you
won’t know the real answer to. They want to know…can he think outside
of the box? They’re nice, but they’re also intense about it. You know this is
no joke when…you’re surrounded by professors and high-ranking army
officers.”
In one case that has become legend, a er being asked about his
hobbies, a candidate spoke during his interview about his love of playing
music. He compared his fascina on with composing music to his love and
interest in physics. He was then asked how he might use both of those
interests to create the perfect sound. The young man described his guitar
in great detail and then explained how he might use a series of
connec ons to amplify the sound. A er the candidate le the room, one of
the commanders conduc ng the interview slapped himself on the head
and said to the other interviewers, “I’ve been trying to build an electric
guitar for my son. He just explained exactly what has been missing!”3
Another candidate’s interview didn’t go quite as well. The lad was asked
if he was a Zionist. He answered, “I love Israel, but I will probably take what
I learn in this program and apply it to professional pursuits outside of the
country.” That man didn’t make the cut. When his father found out, he
marched the boy back to some of the men on the interview commi ee and
told them his teenager didn’t know what he was talking about, and of
course he was a Zionist who wanted to help his country – forever. The
surprised members of the commi ee swore to the father they actually
liked his answer. The kid had been honest. But there were other reasons,
they claimed, why he wasn’t accepted.
Several Talpiot graduates also said they were asked different forma ons
of this ques on: “How many gas sta ons are there in Israel?” Most of them
say they’d es mate the popula on, then divide it into an assump on of
how many cars are on the roads in Israel and they’d try to devise some sort
of logical and mathema cal equa on. But to a man all laughed about it
later, saying they now understand “the commi ee” wasn’t looking for an
accurate answer. They simply wanted to see how poten al recruits acted
under stress and how they thought through problems.
When a young woman going through the tes ng told the commi ee she
spoke Italian, they were impressed, for Italian is not a language many
Israelis learn to speak. Upon hearing this, one member of the commi ee
asked her, “How many people have seen Michelangelo’s statue ‘David’ in
the Galleria dell'Academia in Florence?”
Another candidate who ul mately succeeded and gained acceptance to
Talpiot says, “I was told, ‘Give me the name of a scien st that you look up
to and would like to emulate.’ I thought to myself, don’t pick Einstein, don’t
pick Einstein, don’t pick Einstein…but I panicked and Einstein it was. I then
proceeded to pre y much make up an answer, but the key was to sound
confident and competent. They all probably laughed at me a er I le the
room.”
Another successful candidate found the “mind games” very stressful. He
had been asked, “What is a black hole and how does it work?” He recalls,
“To be honest, I really thought that maybe they didn’t exactly know either.”
While some high school Talpiot candidates regard these psychological
tests and personality profiling s mula ng and others find it stressful, the
personal interview, “the student versus the commi ee,” remains an
invaluable part of the tes ng today. It is a rite of passage and s ll the
cornerstone of Talpiot selec ons.
Yet in the opening years of the program, heavy concentra on on
psychological stamina and academics seems to have been overplayed.
Talpiot recruiters came under cri cism for not finding team players.
One of the most famous Talpiot graduates is a man named Eli Mintz. He
characterizes the early Talpiot people as “very strange. It was twenty
eccentric nerds put together and then told by army officers – who o en
had no idea what we were working on, or how we were doing it – to ‘get
along.’”
“Ge ng along” proved to be a formidable challenge to many of those
super-smart teens. Mintz confesses, “Like many others in the program, I
had come from environments where I thought I was always the smartest.
So when you’re finally in a place where you think, ‘Wow, I’m not the
smartest guy in the room,’ it was great. It was a new challenge. But not
everyone was programmed to think like that, and it led to personality
problems.” He reflects that learning to work with others who are smarter
than he was the most important thing Talpiot had taught him.
Those personality problems would soon be addressed. A er a few years,
Talpiot recruiters added a crucial new phase to the process. They wanted
to see which candidates could work well as a group under challenging
condi ons. To this day, new recruits are measured in this part of the tes ng
by former Talpiot graduates.
Specifically, these tests can consist of working with team members to
come up with a proposal for as many ways to use a bicycle or a shoe they
can think of. Others are asked to design something as a team. Some tests
include using children’s building blocks to construct something. All of this is
happening under ght deadlines, some mes in hot rooms. To add tension
and pressure to the situa on, former Talpiot graduates are lurking behind,
recording every move and every word – or at least the candidates are
made to feel like that’s what’s happening.
By the sixth or seventh year, recrui ng became more formalized.
Professors Yatziv and Dothan knew who they were looking for and how to
find them. And by then there were actual Talpiot graduates with a bigger
say in the program. That was so valuable because their tangible experience
led to prac cal revisions in the Talpiot selec on process.
In fact, a good number of changes took place during the early years in
both the vision and development of Talpiot as its founders struggled with
prac cal concerns and unforeseen problems. In the opening stages of
Talpiot, Professors Yatziv and Dothan weren’t quite sure where they
wanted to take the program, according to Talpiot graduate Amir Schlachet.
“They envisioned a Palo Alto Research Center, similar to the one set up and
developed by Xerox. They wanted to take young people and put them in a
research ins tute and have them think of ways to come up with new
weapons, emphasizing breakthrough technologies. [Ini ally they intended]
that graduates would stay together forever at the research center. But in
the first year, Talpiot commanders realized the idea had to change because
there were simply not enough resources to make it happen in a country as
small as Israel. You can’t just build labs and think tanks in Israel; we don’t
have the resources. So instead they said, ‘We’ll train them and send them
to places where R and D infrastructure already exists, namely in the army,
air force and navy, and also with Israeli defense contractors.’”
For Talpiot to work, it would take resourcefulness, dedica on and the
humility to revise the original plan. And none of these quali es were
beyond its founders, par cularly Felix Dothan, according to Amir Peleg, of
the fi h Talpiot class. He describes the professor as the real driving force
behind the program. “Dothan was very proud of being one of the first to
say that something needs to be done to create a program like this. He was
such a relaxed, nice guy. Not too dogma c, and always genuine. He really
cared about Talpiot, about the army and about Israel – and he did the best
he could to turn his vision into a reality.”
Professor Dothan could not have forged Talpiot without the vigorous
support and coopera on of MAFAT. Because Talpiot and MAFAT are
intertwined, it is important to understand the nature of this important
facet of Israel’s military. Early on when the state was created, Prime
Minister David Ben-Gurion wanted to keep the figh ng men and women
separate from the men and women who would control the money and the
budget. He knew the army would play a large role in the makeup and
development of the na on, but he wanted to separate the guns and the
money in order to maintain a balance of power, and to disable any future
general from controlling both the army and the budget at the same me.
In Israel, the Ministry of Defense is generally operated by civilians who
control the budget for military spending, and the IDF is run by Israel’s
generals.
MAFAT is the Hebrew acronym for Administra on for the Development
of Weapons and Technological Infrastructure. The special department that
was the precursor to MAFAT was just in the beginning stages of working
together with Israel’s defense contractors when the Yom Kippur War broke
out. That new research and development department had li le to do with
the figh ng. During those three hard weeks of war, the head of research
and development, Uzi Eilam, had taken over this crucial unit only a few
weeks before the figh ng began, and like most Israelis, he really had no
idea the war was coming. During the war, he lent out his staffers to
different army units to help in any way they could, yet he retained a core
research and development team in case an assignment came their way.
MAFAT was officially established in the early 1980s by Defense Minister
Ariel Sharon; he envisioned it as an umbrella for all things R&D in the
Israeli military. The new department also had a Foreign Rela ons Unit that
was tasked with buying and selling weapons abroad.
MAFAT’s main goal was and s ll is to develop the Israeli defense industry
– all the Israeli contractors large and small – and get them on one page in
order to work more closely with the Israeli Ministry of Defense.
When Talpiot was founded, MAFAT soon took on leadership and
administra on of the program. Filled with some of the brightest
engineering and military management minds in all of Israel, MAFAT has the
final say on who gets into the program and who doesn’t. Though the air
force is directly responsible for the cadets’ military training on a day-to-day
basis, MAFAT is involved in the educa on of Talpiot students every step of
the way. It helps nurture Talpiot’s young cadets by taking responsibility for
their training and coursework.
Within the organiza on there is a special team known as the steering
commi ee that drives the program from the outside. Headed by the man
or woman serving as Deputy to the Chief Scien st of MAFAT, this
commi ee meets several mes a year to evaluate and nker with Talpiot,
changing or upda ng courses as needed. Talpiot’s commanding officer and
a representa ve of Hebrew University also sit on the steering commi ee.
Some mes a select number of Talpiot graduates are also invited, as well as
other representa ves from Israeli defense contractors or high-ranking
officers in the Israeli ground forces, navy and air force. The commi ee is
also responsible for coming up with strategies to decide where each soldier
will serve a er gradua on.
From its incep on, one of MAFAT’s many missions is to supply Israel with
a never-ending line of highly educated, highly mo vated and like-minded
soldiers who will be er arm and protect Israel. Taking a vital role in the
development and administra on of Talpiot fulfills that mission.

Notes
2 IDF Archives.
3 30th Anniversary Talpiot Yearbook Published in Hebrew.
CHAPTER 4
POLISHING THE PROGRAM

O ne of the many reasons for the overwhelming success of the Talpiot


program is the open mindedness of the officers who run it. Yes, there’s a
bureaucracy. But when it counts, the army knows how to do the job right.
Flexibility and tolerance for trial and error are key in honing a program
that’s part academic, part hard-core warrior training and partly responsible
for developing weapons and intelligence tools of the future. From the start,
the founders of Talpiot were men of science, and they knew that making
mistakes was part of advancement. And the army was quick to realize that
Talpiot’s officers and soldiers needed flexibility to try new things.
Some mes they’d achieve success and some mes they wouldn’t, but it
was clear nobody should fear failure. From the program’s incep on,
Talpiot’s student-soldiers were ins lled with a sense that making a mistake
is perfectly acceptable, as long as you learn from it. That kind of mental
freedom is a prerequisite for true crea vity, and crea vity is the key to
innova on.
As the program matured into its sixth year, it became clear that the
students were some mes more learned in some ways than their
instructors and senior officers. The Ministry of Defense began looking for
someone to take charge who understood the cadets be er. In order to find
the right person, they looked inward.
Ophir Shoham, an outstanding graduate of the second class of Talpiot,
recommended to the IDF and the Ministry of Defense that a former
member of Talpiot was best suited to lead it. They asked him if he had
anyone in mind. He immediately recommended his friend and fellow
member of Talpiot’s second class, Opher Yaron.
Yaron is from Kiryat Bialik, just north of Haifa. A er enlis ng in Talpiot
and ge ng his degree from Hebrew University, Yaron spent five years in
Israel’s communica on corps, improving Israel’s exis ng networks, making
them more flexible, efficient and more secure. When asked, officials at
Israel’s Ministry of Defense would only say that Yaron’s work was
“groundbreaking…and s ll classified,” even three decades later.
Despite the advancements in communica ons that Yaron had
developed, he was s ll unsure about what his next step would be. He
wanted to stay in the army and extend his me serving the country, but he
was looking for a new challenge. When Shoham approached Yaron about
leading Talpiot, he thought, “It’s an interes ng opportunity; I want to do
more than just technical work. I want a chance to work with people rather
than just things.”
Moreover, he strongly agreed with Shoham that Talpiot would be best
served by a graduate who knew the program and the capabili es of the
recruits. “I felt I could relate to them. They were very smart, very confident
in themselves. They think they know everything and that their way is the
right way, so it’s hard for them to accept authority some mes. I was like
that at eighteen, and I’d had the same experiences. So I thought it would
be good for the program for the recruits to relate to a graduate.”
He took over the program in its seventh class in 1985. A er studying
other successful educa onal programs, he was inspired by American Ivy
League schools that not only had great academic programs, but could also
brag about their great tradi ons. The seventh year of a program is too
early to have a true tradi on, but Yaron knew that Talpiot was capable of
establishing a tradi on that would be the envy of other IDF units and
academic programs in Israel. And fostering such tradi on was crucial, he
reflects, because “the early excitement for the program was star ng to
wear off. There was a leveling off of resources for Talpiot. When I came in, I
wanted to reinvigorate the program to create a new tradi on where it
would be known as a unit and a part of the army that was constantly
innova ng.”
Yaron didn’t want to make immediate and drama c changes for the sake
of making change. He believes that’s o en a mistake made by new chief
execu ves in the business world. Instead, he wanted to build on what had
already been established and enhance it.
In the year before Yaron arrived, Talpiot had started recrui ng women.
Six females were dra ed in 1984, none in 1985 and three in the eighth
class. As commander of Talpiot, Yaron never tried to recruit women
differently from the way he recruited young men. “We simply sought the
best people.” The women who were recruited wanted the full program,
though typically high school girls do not want to commit to the army for
ten years. “We quickly found out it was not a problem for the women we
were considering,” he recalls.
For most basic training programs in the Israeli army, men and women
are separated, as they’re usually in single-sex units. But in Talpiot, the men
and women in the same class go through drills equally and together. That
means long runs and difficult hikes, shoo ng courses, parachu ng school,
obstacle courses and beyond.
Marina Gandlin (twenty-sixth class of Talpiot) admits, “I was mostly
nervous about boot camp. But I talked myself into it saying to myself if
other girls can do it, I can. Talpiot is a long program, of course, and a lot of
people don’t make it all the way through. I wasn’t sure if I was going to be
able to make it un l I was in for six or eight months. By that me, I had
built up enough confidence. Thirteen students dropped out of my class.
These were people who were quite capable of doing advanced physics, but
Talpiot was just too difficult for them.”
With Talpiot’s high drop-out rate, keeping female recruits was especially
hard. Yaron recalls that with one undecided female recruit, “I wanted her
to join, but I tried not to pressure. She did join. She was the lone woman in
the class and wound up dropping out. I felt very bad about it.”
He remained the leader of Talpiot for two years before handing over the
reins to another graduate. In fact, from the me Yaron took over – with
one excep on, for a short period of me – only Talpiot graduates have led
the program since. His bold move in accep ng the job set an enduring
precedent.
Two of the program’s most innova ve Talpiot commanders were
Lieutenant Colonel Avi Poleg, Yaron’s direct successor, and Major Amir
Schlachet (a man who would later go on to major success at Israel’s largest
bank, Bank Hapoalim).
Poleg grew up in Haifa in the 1970s. As a teenager, he was quiet, curious
and studious. He was also a musical prodigy, a cellist. Once a week he’d
commute two hours to Tel Aviv for cello lessons. He clearly remembers the
trauma of the Yom Kippur War and always wanted to help his country. But
because of a few health problems, he knew he’d never be able to serve in a
combat role.
When he was asked to apply for Talpiot in 1981, he had never heard of it
before, like most of the enlistees of those years. The program hadn’t been
publicized for it was regarded as a military secret. In Israel, military secrets
are closely guarded, even in families and circles of friendships; people
simply know not to ask too many ques ons. Since nobody had yet
graduated from the program in early 1981, there was nobody to ask about
what happens inside of it. He had to trust his gut.
He became a member of the third class of Talpiot and never looked
back. A er gradua ng from Talpiot’s academic program, he was picked by
the navy to develop electro-op cs for ships, first working on devices
designed to deceive enemy radar and missiles, then on developing infrared
and thermal cameras to detect ships, weapons, missiles and any other kind
of imaginable threat.
A decade a er enlis ng in Talpiot, Poleg became its leader, in charge of
day-to-day opera ons. His fieldwork in the navy helped him get to this
posi on, one he had coveted for some me. A former superior at the
Ministry of Defense said of him: “It was really perfect for Avi. First off, he’s
a sincere patriot. He also has a real understanding of physics. And perhaps,
most important was his interest in the field of educa on.”
During the nine es, Israel’s military technology was moving forward at a
rapid pace and Talpiot students and graduates were playing an outsized
role in developing those technologies. Many of them, including cellular
phone technology and data encryp on, would soon have major civilian
uses.
As the army’s needs were changing, Talpiot’s officers had to make
adjustments to the list of desirable quali es in candidates. Teamwork was
becoming a bigger part of the equa on because different kinds of systems
had to be integrated into different units. Each year, picking the right twenty
or twenty-five people was becoming more and more important.
“Suddenly, we had to start looking for a new combina on of a ributes,”
Poleg said. In addi on to high cogni ve scores and scien fic thinking, they
were looking for people who could lead: officer tes ng and personality
exams became crucial. Poleg revamped two parts of the exam – the group
test and the interview.
“I wanted to check mo va on, moral value and, of course, personality.
We would run intense social simula ons in which the candidate was put
into a high-pressure leadership posi on. How do you try to mo vate your
classmates who might be falling behind? How do you deal with those who
refuse to take part in a certain project or ac vity? My goal here was to see
how candidates coped with social issues, leadership issues and paying
a en on to everyone. I needed to be confident that the candidates I
picked would be crea ve, intelligent, inven ve, with the ability to move
from one area to another, and be able to take leadership in a group while
being part of that group. It was also cri cal to get a sense of how cadets
might act when having to deal with someone above them and below them.
Finally, I also needed a sense of their moral values and willingness to make
a contribu on to their country and society. I was always confident I could
move the right students forward, but they had to have the roots. The trick
was dis nguishing who did and who didn’t.”
In the commi ee interview, Poleg might say, “Tell me something
interes ng that you saw last month that you didn’t know, what you learned
about it and how you increased your knowledge. Maybe an interes ng
instrument, an interes ng science program you watched on television, an
interes ng ar cle you read about science.” This was the star ng point, and
he would use it to find out the level of the candidate’s curiosity and how
far he’d go to sa sfy that curiosity. “I was looking to see if the candidate
made a real effort to inves gate.”
Poleg also refined the process of assessing the candidate’s ability to
think. Famously, he would give them a sophis cated ar cle from a science
journal about something he was confident the recruit didn’t understand
and had not studied. The recruit would quickly read the ar cle and then
was asked a series of ques ons about it. The goal was not to test him on
his knowledge, but to observe his thinking process.
Poleg believed that an incisive commi ee interview gave him the best
sense of the candidate, well beyond test scores. In one memorable
instance, a candidate had not done par cularly well on the previous
tes ng, but Poleg gave him a simula on. “He started to flourish…he was so
enthusias c and really animated. ‘I would do this and I would do that…’ It
was as if he were suddenly conduc ng an orchestra; as if he had found his
voice right then and there in front of the commi ee. This was exactly the
answer I was looking for. He has it! I viewed the commi ee in part as a
trainer to pull something out of a candidate, and I used similar methods as
a commander and educator once those poten al recruits were in the
program.”
Of course, there is no foolproof way of picking the best cadets for
Talpiot. As Poleg had projected, social tes ng took on increased
importance in the late 1990s and earliest years of the twenty-first century.
It is generally accepted that no Talpiot commander should stay in the
same posi on for too long. By 2003, the ming was perfect for new ideas
and new leadership, and Amir Schlachet became commander of Talpiot.
By the me Schlachet was planning his army career at the age of sixteen,
the veil of secrecy over Talpiot was star ng to li and the program had
become known throughout the country. Soon it became a top priority for
any young Israeli who was interested in physics, science and math and was
able to achieve a high academic status. In many ways, students who target
admission to Talpiot are akin to students in the United States who want to
go to Harvard, MIT, Princeton or Yale.
At the age of seventeen, joining Talpiot wasn’t exactly Schlachet’s life
goal. But as he made it further and further through the labyrinth of exams,
he became more and more interested and more hopeful that he’d make
the cut. “I knew I wanted to do cool stuff; I knew engineers and physicists. I
didn’t really know what research and development was, but I liked the idea
of merging science and defense.”
A er his three years of course work, he graduated with his degree in
science and physics from Hebrew University. He was placed in an air force
research and development unit by the Talpiot commanders and Ministry of
Defense, and worked in a special unit developing airborne electronic
systems (used in communica ons, radar and air-to-air and air-to-ground
targe ng).
As he was finishing his Talpiot service, he was ready to pursue an
advanced degree and then move to the world of business. But an Israeli
Ministry of Defense legend (and Talpiot graduate), Eviatar Matania, a man
who would later be known as “the right hand of Talpiot,” asked Schlachet
to stay. Matania recognized something extra special in Amir Schlachet: he
saw a man who understood the need for Israel to be ahead of the curve in
technology, a great project manager and an officer who could get everyone
together to push toward a common goal. Matania offered him the chance
to command the en re day-to-day program and to revitalize it.
Schlachet immediately accepted and put his business career on hold,
indefinitely. With ruthless honesty, he assessed every feature of Talpiot,
“looking at the en re program from above. If something added value, we
wanted to strengthen it. If it was weak, we wanted to drop it. We changed
the so ware for aggregate numbers-crunching. Talpiot keeps evolving and
to a large extent; so does the screening process. One of Talpiot’s biggest
strengths is that we do our own screening, so we wanted to improve and
intensify that. We redesigned the en re process. We are never on lock
down and that’s a tremendous strength.”
Breathing new life into the program, Schlachet upgraded the posi on of
top officer to a Talpiot graduate and gave him more responsibility – he
stays for a few years, not just one year, before moving on to something or
shipping out from the army. “We didn’t want to invest extra training in
future leaders and then have them leave a er just one year,” Schlachet
explains.
While s ll heading Talpiot, Schlachet persuaded a former class
commander to stay a er his me in the army was up. Dror Ben Eliezer was
given a three-year role as head of admissions under Schlachet. Dror is one
of the few Talpiot graduates who also had a brother in the program, Barak.
They had grown up inside the Old City walls of Jerusalem, playing American
football with yeshivah students from the United States. Barak would later
use many of the management techniques he learned in Talpiot and apply
them to Israel’s police force.
Schlachet is a very humble man, as are most Talpiot graduates. He
begrudgingly admits that colleagues and other Talpiot graduates might be
right when they said, “Amir breathed new life into the program, revamped
it and made it be er.” What he is most proud of is that he revolu onized
the screening process, which was in a transi onal phase at the me. As the
tes ng lted increasingly toward social and personality factors, he actually
came up with a be er way to find out who is right and who is not right for
the program. The “quiet tests” (where candidates don’t know what they’re
being tested for, or that they’re being tested at all) reveal a good deal,
thanks to psychometric advisors. These are third-year Talpiot students,
(chosen on the basis of commanders’ recommenda ons) and Talpiot
graduates who have gone through special cer fica on to par cipate.
Marina Gandlin became a proctor for the social tests given to
prospec ve Talpiot cadets. She describes the group tests that are given
over two days. “We split them into teams and ask them to build something
out of paper or out of blocks, to see how they interact with other people. I
don’t want someone telling the other people what to do and what not to
do. I don’t want people who are rude or too pushy. No violence and no
shou ng. Those are definitely signs you’re not right for Talpiot. I know it
sounds funny, but you can easily tell who’s good and comfortable and who
has ideas about how to get the project completed, and also who can get
others on his or her side to complete that project. We watch them move
forward with their projects right in front of them. There are no two-way
mirrors or anything like that, just ten or twelve people in a room working,
and two or three Talpiot graduates watching them to see how they
behave.”
When asked what makes Talpiot so unique when compared with
engineering or physics programs in other armies, Schlachet gives three
reasons: 1) the fine-tuned selec on process; 2) unique training, both
academic and military, with emphasis on the big picture; and 3) success in
finding the right posi on for the graduate to serve.
This last factor is cri cal, both to the Talpiot graduate and to the country.
Talpiot officers work to build job descrip ons for posi ons in research and
development that are high on the priority lists of the army, navy and air
force. “We want them to be interested in the things they’ll be doing a er
gradua on,” says Schlachet, “because, hopefully, they’ll be doing those
jobs for at least five years. If they’re interested, they’ll be even more
mo vated. The solu ons we’re working on for the IDF aren’t trivial, for
some the work will be the difference between life and death. We work on
placement of each graduate during their en re three years of study. Our
goal is to put each Talpiot graduate in a posi on where he or she will have
maximum impact.”
Schlachet points out that one of the strengths of the Talpiot experience
is that it is collabora ve. “Somehow, it is not compe ve in any way. In fact
it is one of the least compe ve places I’ve ever worked in. People help
others even if it means they get a lower grade or do less on something. It’s
fantas c. Very close es are built, maintained and kept forever.”
Despite the fact that Talpiot graduates have taken responsibility of day-
to-day opera ons of the program, there are s ll people above them from
Israel’s Ministry of Defense, par cularly insiders at MAFAT, the Israel
Defense Force’s research and development arm. Throughout the selec on
process, administra on of the program and placement of graduates, these
career army officers make sure Talpiot gets what it needs and that Talpiot
gives back what the army needs from them.
A high-ranking member of the Israeli Ministry of Defense (who would
not allow his name to be used) pointed out that “the unique thing about
Talpiot is that it con nues to reinvent itself, stay ahead of the mes and
improve itself. Part of that is because the graduates care so much; and
they’re somehow constantly able to leave something be er behind. It
would be great if the rest of the country were like that as well!”
CHAPTER 5
IT ALL STARTS IN HIGH SCHOOL

O nce the Talpiot program got off the ground and had refined its selec on
procedures, it began to make a name for itself as an elite unit. Today,
thirty-six years a er the program began, the IDF’s website describes the
exclusivity of Talpiot: “The program accepts 50 outstanding students from
those in the science tracks in high schools.” Compe on for admission to
the program is fierce.
Ron Berman is an uber-student who not only was accepted to Talpiot’s
nineteenth class, but also has been educated in Denmark, Tel Aviv
University, the University of Pennsylvania’s Wharton School, and is
currently studying for his PhD at the University of California, Berkeley. This
student of students advises anyone trying to get into Talpiot: “It all starts in
high school.”
Many of Israel’s high schools are actually dedicated to helping their
students get into elite technology and “thinking” units like Talpiot. One
ini a ve in Jerusalem was designed specifically with Talpiot in mind. The
Israel Center for Excellence through Educa on is run by former Talpiot
commander Avi Poleg. He uses many of the techniques he perfected at
Talpiot and shares them with schools run by the state, throughout Israel.
The Center is integral in se ng the curriculum and helping to run one of
the few boarding schools in the State of Israel, the Israel Arts and Science
Academy. The school shares its campus with The Center.
Students admi ed to this elite program are already extremely talented.
But the school is designed to help them focus on mathema cs, chemistry,
physics and computer science at a higher level than they could get at a
state-run school anywhere else in Israel. Some students commute from the
Jerusalem area, but most live, sleep and eat in dorms. The teenagers at the
school come from more than a hundred diverse communi es in Israel.
They come from large ci es and small towns, from kibbutzim and
agricultural communi es, and the program is open to both Israeli Jews and
Arabs.
The Israel Arts and Science Academy and The Israel Center for Excellence
Through Educa on give students the tools they need to learn be er and
faster, just like Talpiot cadets. Poleg’s teaching methods are the same as
they were in Talpiot. “We don’t direct students to a certain point; we want
to prep them with the abili es and values and talent to cope later on with
challenges. Talpiot’s philosophy of how to promote independent thinking,
curiosity and mo va on is relevant here.”
When asked, “How do you implant curiosity and a desire to learn in a
student?” Poleg answers: “You’re asking for the whole Torah on one leg, as
Rabbi Hillel once said.” [The famous sage was once asked if he could
summarize the en re Torah while the ques oner “stands on one leg.”] “In a
nutshell, it is developing study units that will pull students into an
adventure. We don’t start by saying today we are going to learn about this
or that. We start with a legend or story. We try to put the student in a
posi on that simulates the situa on. We tell them, ‘This me you’ll be a
historian; a scien st; a detec ve.’ Students should learn under different
hats to keep it interes ng. It is a way to build tasks in a moderate way,
pu ng much of the responsibility in the hands of the students themselves.
The teachers are not supposed to be the source of knowledge but rather
facilitators of the learning process. That means they are there to guide the
process and instruct slightly, to let the students come up with their own
conclusions; to let them fail and not correct them too early. The teachers
should ask many ques ons, but provide few answers. Answers should
come from the students.”
There are clear differences between a boarding school and a military
base. While the end goal is the same, implementa on isn’t iden cal. But
Talpiot rules and methods do hold true. Poleg sees it as his responsibility to
“strengthen independent abili es and self-esteem. I deeply believe that
once you teach a learner you will harvest the results later on.”
Some might call the Israel Arts and Science Academy a “prep school for
Talpiot.” Admission to Talpiot is always in the minds of the program’s
students and teachers, and they’ve sent more than their fair share on to
Talpiot success.
Nachshon, a similar program for high school students, later became
prominent. It was founded by Eviatar Matania of Talpiot’s sixth class. The
program is named for the valiant biblical figure Nachshon ben Aminadav,
the first Israelite to jump into the Red Sea before the waters parted. While
Nachshon’s students are not necessarily Talpiot candidates, several cadets
have come from the pres gious program.
In Israel, many of the top high schools align themselves with top
universi es, either formally or informally. One of Talpiot’s earliest “feeder
high schools” was a school called Handassa’eem. Handassa means
engineering and geometry in Hebrew. It used to be closely affiliated with
Tel Aviv University and has sent Talpiot some of its finest graduates,
including Eli Mintz. He recalls, “I was recruited to Talpiot a er the army
received a recommenda on from the principal of my high school at
Handassa’eem. Five people from my high school class went on to Talpiot.
Overall, Handassa’eem has sent a high percentage.”
Another prominent graduate who went on to Talpiot is Ophir Kra-Oz.
The principal at the me was Yohannan Eilat, who “turned it into a tech
powerhouse. He used to take a map of Silicon Valley and put it over Haifa
and tell everyone, ‘look – a perfect fit.’ That was in 1988, long before
Silicon Valley was something everybody in the world knew about. He was a
real visionary.”
While Kra-Oz was in high school, he and many other future Talpiot
students were so far ahead of their instructors in computer science, they
wound up teaching the classes. As the world adapted to computer-based
life, teachers had trouble keeping up at first. It quickly became clear the
de was changing throughout the globe as students began teaching the
teachers.
Yet a big advantage emerged at Handassa’eem in those years. Many
Russian immigrants with very sophis cated educa on couldn’t find work at
their level in Israel; and so, many became teachers at Handassa’eem.
The school has since moved and now stands in the town of Herzliyah,
just north of Tel Aviv. The principal today is Orit Rozen. Handassa’eem
furnishes so many recruits to top army technological units “because of the
mul tude of projects the school offers, which is unprecedented in the
country and abroad. There is a diversity and high level of scien fic work,
including a combina on of various disciplines – computer science,
technology, medical science and more.” Today “more” includes robo cs
and facili es for students showing an early acumen for satellites,
aerospace and bio-technology.
In Haifa, not far from the Technion, is a high school named Leo Baeck.
Students are accepted from all walks of Israeli life. Many pay some tui on,
though about 10 percent are on full scholarship. There are forms of
assistance for other students as well.
The mission of the school is to provide a pluralis c educa on for its
students who come to the school from all over the northern half of the
country. There are about a thousand students and 150 teachers, a ra o just
about any school in the world would envy. Leo Baeck set a modern-day
record in 2005 by sending five graduates to Talpiot including Marina
Gandlin, who later became an early pioneer in short-range missile defense
and the Iron Dome.
In Jerusalem, a pres gious public high school also known for supplying
Talpiot with numerous cadets is known simply as L’yada, which translates
to “next to” in English. It gets its name from being “next to” Hebrew
University.
Israel is a country where kids have to grow up fast so they can contribute
to the security and well-being of the state early on. And the Israeli high
school system is actually so important that it has become a major source of
fundraising overseas through programs like “Friends of Israel Sci-Tech
Schools.” Seventy-three high schools fit into this special category, where
the focus is on robo cs, engineering, nanotechnology, biomedical
engineering, aerospace and computer science. These schools have not only
educated scores of Talpiot cadets, hundreds of their other graduates have
gone on to provide important security solu ons and become Israel’s high-
tech business leaders.
A strong spirit of giving back permeates Israeli society. Many execu ves
who have a ained the highest levels of Israel’s corporate hierarchy have
established ways of helping the country, its students and future leaders.
Former Israeli President Shimon Peres has ini ated one such project. Over
the last several years, the popular statesman has been helping Israeli high-
tech CEOs meet and mentor the na on’s most promising high school
students through a program he helped found with the charitable
organiza on, the Rashi Founda on (designed to encourage Israel’s teenage
inventors). He has enlisted the help and support of many of Israel’s biggest
corporate leaders, including Sammy Segol who runs Keter Plas c, one of
Israel’s biggest companies and biggest exporters. Segol encourages his
employees to get involved in mentoring future leaders as well.
One of those employees is Barak Ben-Eliezer, a man who could not
possibly be more focused on be ering the state of Israel. (Selected for the
Talpiot program in 1992, he was part of the famous fourteenth class whose
members founded the storage company XIV and sold it to IBM for $300
million.) Two of Ben-Eliezer’s Talpiot connec ons have also dedicated a
good part of their me to helping young, enterprising minds. Uri Rokni
develops algorithms for an Israeli company called Mobileye. (He’s working
on greatly improving the safety of driving by pu ng high-tech an -crash
systems in the cars we drive. The company listed on the Nasdaq Stock
Market in August 2014 and the stock quickly soared, becoming the darling
of Wall Street.) In his spare me, he volunteers to help increase the level of
competency in math in Israel’s high schools.
Uri Barenholz graduated with Ben-Eliezer’s fourteenth class of Talpiot.
He went from data storage to biological engineering research at the world
renowned Weizmann Ins tute in Rehovot, Israel. For fun, he teaches
physics at a junior high school in Holon, a suburb of Tel Aviv.
Some of the students he teaches, like the talented young scien sts being
graduated from specialized high schools, will set their sights on advanced
university degrees. Many will aim for admission to Talpiot, for they know
that it is the stepping-stone to the best university educa on, the highest
army accolades and a promising future.
CHAPTER 6
THE WORLD’S FASTEST LEARNING CURVE

Fneedromtotheoutsource
start, the Israeli army and founders of Talpiot knew that they’d
some of the educa onal aspects of the program to
Hebrew University.
Most Israelis don’t start at universi es un l they are at least twenty-two
years old, four or five years later than most students in the United States.
Most Israelis are discharged from regular army duty and then go to see the
world. They go to India. They trek through Nepal. They escape to Thailand.
In fact, there are so many Israelis in parts of those countries, street signs,
store and hotel markers are in Hebrew. One shopkeeper in India was
shocked to find out there were only about six and a half million Israelis. He
thought that since his town was constantly overrun with men and women
from the Jewish state, there must be hundreds of millions of them. Other
young Israelis go to South America for months at a me, hiking through the
Andes. Many go to both Asia and South America.
When they return to Israel, they may register in one of Israel’s nine
universi es, many of which are world renowned, including Tel Aviv
University, Ben-Gurion University of the Negev, the Technion in Haifa, and,
of course, Jerusalem’s Hebrew University. Others might go to one of
several dozen colleges in Israel.
But because Talpiot cadets serve at least nine years in the military, they
immediately begin studying for their academic degrees at the age of
eighteen, when they enlist. When they finish their coursework at Hebrew
University, they have their bachelor’s degree in mathema cs, physics
and/or computer science. This advantage gives them peace of mind,
knowing that they won’t have to start their studies a er they’re out of the
military.
When the army is paying for you to study, however, you don’t have the
luxury of falling behind. In Talpiot, if you do drop back, you’ll get kicked
out.
Speed has always been an important part of the program. Because
Talpiot students get fewer weeks to study than their university
counterparts, the academic program moves faster. One reason for that is
simply because the cadets are in the army and they have other things to
do. Another reason is that the army is inten onally training the cadets in
how to think faster.
There is no magic to making a student learn faster. The way it is done is
to emphasize group learning. The thinking is that if you’re with other
cadets in a military-like se ng twenty-four hours a day, seven days a week,
you bond. When part of the group moves faster, the rest of the group will
keep the pace.
The speed of the coursework is much quicker than at a regular
university. The cadets train and learn as a class. Academic compe on is
not part of the program and there is no chea ng in Talpiot. Many of the
professors allow students to share work, as they encourage cadets to help
one another. The thinking is that as each cadet brings different kinds of
strengths from different backgrounds, integra on is greatly encouraged.
That emphasis on teamwork helps create high levels and higher speeds for
development and for learning the course materials.
But some mes that speed can be a problem and the 25 percent dropout
rate a ests to the challenge. Even some of the top Talpiot recruits who
went on to become some of the most successful Israelis of all me have
complained Talpiot’s coursework moves too fast.
Marius Nacht is a co-founder of Israel-based Check Point So ware
Technologies. Their Internet protec on so ware defends almost all of the
companies in the Fortune 500 from web-based a acks. Nacht is a graduate
of the second class of Talpiot. He was born in Romania while his parents
waited anxiously for the Romanian government to grant their family exit
visas. In the 1960s, Romania held its Jewish ci zens hostage. If they wanted
to leave, the Jewish Federa on of North America had to cough up a $5,000
ransom for every exit visa. His parents had started the immigra on process
a decade before the paperwork finally came through.
Nacht was three years old at the me, and does not recall his first days
in Israel. But he does remember growing up in a rough, industrialized part
of the coastal town of Ashkelon. He says his family’s situa on gradually
improved to the point where they were eking out a middle-class existence.
Back then, standardized tes ng wasn’t exactly part of the norm, so Nacht’s
family didn’t realize Marius had a special academic gi – and neither did
Marius.
His father insisted that he a end a voca onal high school, a place where
he could learn a trade. Marius a ended ORT, one of many programs
funded by the global Jewish community. He says, “I wasn’t interested in it. I
was doing what he told me to do. We studied many things, including
electronics.”
In 1980 an army recruiter came around looking for the brightest
students. It was rare for the army to look beyond the established high
schools in the major popula on centers of Jerusalem, Tel Aviv and Haifa.
And it was even rarer, in those early days, for the Talpiot program to recruit
someone from outside of those areas. But from Marius’s class, two
students were selected for Talpiot tes ng.
Marius was intrigued by the exams. Acceptance in Talpiot meant more to
him than just the opportunity to be part of this new and exci ng part of
the Israeli army. It meant he had been absorbed by the country he moved
to as a boy; that his intellect had been recognized, though he came from
the depressed town of Ashkelon, a town o en ignored by Israel’s
established elite in Jerusalem, Tel Aviv and Haifa.
But once he was in the program, he wanted out. “The other guys were
very, very smart in terms of math and physics, and I was not at the top of
the class as I had been in high school. And it was more compe ve than I
expected. The five top guys would elbow the others. ‘Come on, why are
you asking that stupid ques on? The professor just said it five minutes ago,
why are you asking again?’ We knew that the first class, the year before
ours, started with thirty students, but a year later only twenty were le . So
I was sure I was going to be kicked out. I finished the first term with a lousy
academic average of 65. To me it was evidence that I should drop out – and
I wanted to. Why con nue? I was just prolonging my military service
instead of doing the things I really wanted to do.
“So I went to Hanoch Tzadik, the guy that you have to talk to if you want
to leave the program. He was a psychologist. I explained to him that I’m
not that good. I was ge ng a lot of homework and not even ge ng half of
it done by one in the morning.” Tzadik (who later became one of Israel’s
best known professors of psychology and a mo va onal coach to
execu ves) convinced him not to quit. He told Nacht he had put a lot of
tension and pressure on himself and that’s why he couldn’t concentrate.
He made him promise that every other day he would run around the
campus, five or six miles, a minimum of three mes a week. Nacht recalls,
“Because I was much cooler about it and not pressing myself, my average
grades jumped from 65 to 85. I figured if that’s the trend, I might even
finish with a reasonable average – and I stayed on. Hanoch Tzadik was a
very important person in my life and obviously made a huge impact.”
Tzadik is an appropriate name for a psychologist who helped so many
people that would later become crucial to Israel’s security. The word tzadik
in Hebrew means “righteous.” It is o en a tle some mes given to biblical
figures. In short, a tzadik is someone who lives by his faith.
Hanoch Tzadik notes that for most of the Talpiot students, it was the first
me in their lives they needed help and to some it was a real crisis. “Their
main problem was dealing with difficulty, not the course work.”
“It was my main job to help them,” he affirms, but there was no blanket-
solu on for each cadet with a problem. “I didn’t really tell them anything
at first, I just listened to them. I had to make them believe they will
overcome the problems, and it’s a very personal thing. I really believed
that most of them could overcome it. The ones that le generally did very
well later, but it wasn’t the right me for them. These were not failures.
They just weren’t ready for this kind of thing. “
Ge ng through Talpiot was never easy, even for those who thrived on
the challenge. One such person is Doctor Aviv Tu nauer, one of a few
Talpiot graduates to go to medical school a er finishing his army service.
He’s an anesthesiologist specializing in pediatric surgery, and he agreed to
be interviewed on a busy day of surgery. We meet in the hospital lobby at
Hadassah Medical Center and we talk in the locker room as he puts on his
opera ng room scrubs. So that he’d never forget it, Tu nauer sets his
locker combina on to the number represen ng a certain isotope of
uranium that can sustain a fission chain reac on. (For him, that’s a
memorable figure!)
He explains what would happen during surgery, the goal of the
opera on and his role. The surgery will be on a two-year-old boy who has
an ar ficial heart valve needing repair. It will be Dr. Tu naeur’s job to
sedate the child.
Around us, doctors are listening to our interview. The cardiothoracic
surgeon stops his pre-opera on procedures for a moment and looks at
Doctor Tu nauer. He asks in Hebrew, “Who is the person doing the
interview and what’s it for?” Tu nauer replies, “It has to do with my
Talpiot experience.” The surgeon asks in alarm, “Isn’t that all top secret?”
Tu nauer chuckles and the interview con nues.
He tells me that in addi on to going through the program, he also
served as a commander of the fi eenth class of Talpiot. He didn’t quite
realize it when he was a cadet – but it hit him as a commander – that
educa ng recruits straight out of high school has advantages. “At that age,
you’re not responsible for a family; no kids, no jobs. You can study un l one
or two in the morning if need be, and it o en is need be. The army gets a
class of students who are free and able to learn.”
The other side of the coin is that, as youngsters, they have constant
complaints. As a cadet, Tu nauer complained about the same things he’d
later have to address as a commander. “We complained that the lecturers
go too fast, and it’s unfair because we are tested on more than our
counterparts in the university. We cover 30 percent more. And the
response was always that the lecturer will go as fast as his class allows him;
if you understand everything, he will go on. We would complain pre y
freely about things. We were very cynical.”
Before the sixth Talpiot class, the commanders of the program had not
graduated from the program and they weren’t completely in tune with this
new breed of intellectual super soldier. But even once the commanders
star ng coming from the Talpiot ranks, as Tu nauer did, there was s ll
tension between adolescent and adult, student and teacher. “As a
commander, I had difficult moments with my cadets, but it was much
easier for me than for previous commanders. You simply understand the
dynamics because you witnessed them. They would complain daily, weekly,
about the classes, the learning materials, the extra-curricular plan, the
quality of food, quality of rooms, the cleanliness of rooms, the burden of
guarding the building, flaws in building security – whatever, you name it. As
a commander, you know there are recurring themes. That’s the way it is;
that’s the way it’s always been.”
The cadets would ask whiny ques ons like any other teenager. “Why are
we guarding the building?” Tu nauer would answer dully, “Because we are
soldiers and that’s what soldiers do. You’re given two hours a week of
guard duty. That doesn’t hamper your studying. The issue is closed.”
As we finish speaking, Tu nauer washes, disinfects, then strides into the
opera ng room. A small pa ent whose life lies in his hands is wai ng.
The doctor’s descrip on of the Talpiot training reflects the goals set for
each of the three years of the program. Would you be able to handle the
rigors of Talpiot? Here are the expecta ons Talpiot has for its cadets:
First Year. Goal: Build founda ons for resolving problems by learning
advanced mathema cs, physics and computer science.
Basic training period of eleven to twelve weeks, followed by two
semesters of studies las ng up to thirty-four weeks.
Five to six addi onal weeks of military orienta on, visi ng the
different units and branches of the IDF.
Comple on of an officer training course.
Second Year. Goal: Reach a high level of ap tude in math, physics and
computer science. (Almost a third of Talpiot graduates earn a degree in
computer science.)
Thirty-six weeks of studies.
Up to three months visi ng various branches of the IDF to learn more
about their problems and their need for solu ons.
Rigorous paratrooper training.
Third Year. Goal: Bring all educa on and training together; sharpen
leadership and academic skills. This includes a broad range of courses in
the sciences including electronics, aerodynamics and system
authen ca on, as well as military technology.
Acquire a solid background in military engineering, radar, antennas
and military communica on.
Take broader range humani es and social science classes at Hebrew
University, including history, art history, philosophy, Jewish thought
and Arabic studies.
Decide on a discipline and an exper se.
Interview and audi on for posts within the Israel Defense Forces.
“The project” spans all three years of Talpiot training. A few mes a year,
they’re asked to develop and then present a project that solves a problem
in the na onal defense spectrum. In essence, it’s a warm-up exercise,
designed to teach them all of the rigors and stages they will later
encounter when trying to solve real-life defense dilemmas.
For “the project” they come up with an idea that solves a defense
problem, create a budget for it and then produce it. They present their
problem and the way they’ve solved it to a group of army officers who are
brought in to judge and discuss the projects. On several occasions, officers
have been impressed enough with a certain project that they decided to
actually develop it. In addi on, some mes producing a project leads to a
post-army appointment for a Talpiot graduate.
Over the years, it’s become commonplace for second-year students to
introduce themselves to the IDF’s top brass by working on these problems
and the solu ons. Past projects, which will be discussed later, include an
early mock-up of the Iron Dome short-range missile defense shield that has
been remarkably effec ve in knocking missiles out of the air before they
reach their targets inside Israel. Another innova on, the Trophy – a tank-
mounted device that automa cally fires at an incoming projec le in order
to protect the crew inside – had its origins in the Talpiot program.
All Talpiot classes are assigned advisors to help them through the
program from beginning to end. It is the advisor’s duty to serve as a
contact person and liaison between the students, the army and the
university professors. At the beginning of the program, founder Felix
Dothan served as an advisor, a role later carried on by various professors at
Hebrew University. The heads of Hebrew University’s mathema cs, physics
and computer science programs also take on outsized roles in advising the
cadets and serve as a go-between with the army. Hebrew University deans
and rectors have also been integral in the program from its onset. When
those three years come to an end, the cadet gets a promo on as well as
that coveted degree in physics, math, computer science or all three. A er
gradua ng, most Talpiots will then con nue their formal educa on. Many
con nue to study at Hebrew University while doing their army service over
the next six years. The Weizmann Ins tute of Science is another popular
des na on. Talpiots who are accepted there o en study for masters or
doctorate degrees in biology and complex physics. On one floor of
Weizmann, Talpiot students have taken over a line of offices where they
are studying and experimen ng with biotechnology, gene cs and bio-
pharm. A third choice for many Talpiot students is Tel Aviv University,
where they study advanced engineering and business administra on.
The Israeli army has always prided itself on being an equal opportunity
employer for men and women. As noted in an earlier chapter, Talpiot was a
rare excep on when the program began and women were not recruited for
the first several years. But by the me the twenty-fourth class was
assembled in 2003, the evidence was clear – the women had arrived and
had fully integrated themselves in Israel’s most pres gious military
program. Eleven young women were accepted into the program that year.
There even have been several Talpiot marriages. Mazal tov!
CHAPTER 7
TRAINING TO THINK FAR BEYOND THE BOX

M atan Arazi’s father was an Israeli diplomat and Matan lived for part of
his youth in Japan. He was closely connected with other westerners who
a ended “The American School in Japan” in the 1980s.
At this me, many of the fathers of the students who a ended “The
American School” worked for the big American banks and brokerage firms
in Tokyo. Word got around that Matan was a computer whiz. One day, he
got a call from the father of one of his friends who worked for Morgan
Stanley. He needed help coming up with a system that could transfer funds
and stock orders quickly across phone lines. Matan, using secure
communica on lines provided by Morgan Stanley, was able to develop a
so ware program that could transfer money and stock transac ons
instantly across the world to other Morgan Stanley offices. Now every
brokerage firm in the Western world has that technology, but Matan was
about fi een years ahead of the rest of the world. He was just fourteen
years old at the me. He went on to do consul ng work for Goldman Sachs
in Tokyo.
Back in Israel, the army realized that Matan had an enormous amount of
experience working in fields needed by the military. Talpiot was quick to
accept him and train him further for the army’s use.
“The most amazing thing about Talpiot,” says Matan, “is that the tools
you learn to use can really help you make a 1 percent difference in the
ba lefield. Think about that. A 1 percent difference. An infantry man can’t
make a 1 percent difference. Maybe a pilot in a small engagement
concerning a major target can make that kind of difference, but Talpiots are
doing it constantly, day by day, on many different projects. In many ways,
we can be the difference between life and death for hundreds, or even
thousands, of people.”
To make a contribu on like that, you first have to be confident that you
can – and that it’s a possible thing to do, even if certain tasks seem
impossible. “You can do anything” is ins lled in Talpiot recruits. “And if you
can’t do it,” chuckles Matan, “you know that another Talpiot graduate
either has, or is on the verge of doing it. Nothing is impossible.”
Talpiots are taught from the first day of induc on that they can do
anything they put their minds to. How does the program inject the cadets
with such confidence?
Ra’anan Geffen was an early Talpiot graduate from the third class,
inducted into the IDF in 1981. “Talpiot ins lls you with a confidence you
won’t find anywhere else, but you s ll have to define yourself,” he says.
“Part of that is through actual fieldwork, but the building blocks are being
laid when you’re doing difficult, but important, course work. Even if you
don’t understand it immediately, the material is taught in a prac cal
manner, allowing even the trailers to understand it – maybe a li le later,
but regardless, they get it, and that gives you a high level of confidence.”
Graduates are some mes given mul -million-dollar budgets in the first
months of their research and development tours of duty to create and
improve the IDF’s weapons arsenal. This vast amount of responsibility, and
the help they’re given from more experienced designers and programmers,
o en puts them in a posi on where they’ll have at least some success. And
the more success they have, the more they believe in themselves. They’re
also taught from day one that if they don’t have the answer, being in
Talpiot gives them a rare advantage. They will be within one or two
degrees of finding another Talpiot graduate that does know the answer to
the problem they’re working on, or at least has answers that could lead to
a solu on.
Talpiot was not meant to be a machine that churns out like-minded
thinkers. The founders envisioned a program that would give their cadets a
founda on to do whatever they wanted to do. The program was designed
to breed crea vity, not conformity. And no two Talpiots come out of the
program the same. The program has a life-long influence on the graduates
who have gone on to do many different, but equally mind-boggling things.
Talpiot is also geared to help force recruits to work and think both within
a system and outside of that system. It also strives to provide cadets with
the ability to lead as many graduates will be managing and working closely
with some of Israel’s most inven ve engineers, avionics experts, computer
programmers and intelligence analysts.
One way of training a student to think is to play against his strengths,
says one insider. “Don’t let him rely on methods of learning or problem
solving he’s already used to. If you force that person to learn a different
way, you’re forcing him to think a different way.”
Tools used to find a student’s hidden poten al include an overload of
training, forcing students to work and study together, and pairing Talpiots
with current and former class commanders. Other key factors in the future
success of its graduates (in and out of the Israeli military industrial
complex) are informal instruc on in me management and Talpiot’s
proficiency in developing the student’s ability to dis nguish what’s
important from what isn’t to achieve a given objec ve.
Talpiots are given access to something most soldiers don’t have un l
they’re older and more experienced: informa on about how things work.
In the broader military, a soldier is a soldier, and he operates in a need-to-
know environment. The goal of Talpiot is to expand students’ knowledge so
they’re more in the loop about what happens behind that green curtain
separa ng commanders from the men and women who rank below them.
The refining of that great founda on came in large part from a man who
once headed MAFAT, General Yitzhak Ben-Israel. He was born in Israel in
1949, toward the end of the War of Independence. Had Talpiot been an
op on for him when he was eighteen, he would’ve been a perfect
candidate. General Ben-Israel has the intellect of a rocket scien st, the
brawn of a special-forces commander and the confidence of a true military
leader.
His studies focused on mathema cs and physics (and philosophy, for
fun). In essence, he was a Talpiot before Talpiot existed. His resume is that
of a true unsung hero, a man who spent his life in the army without being
no ced by the public at-large.
Ben-Israel joined the Israeli Air Force right a er the Six-Day War in 1967.
Before taking leadership of MAFAT, he held high-level posi ons in the
intelligence and weapons development units of the IAF. He was in charge
of the Israeli Air Force’s Opera ons Research Branch. He has won Israel’s
pres gious “Security Award” twice for developing s ll-classified security
systems. The first me Yitzhak Ben-Israel won the prize was in 1972 when
he was just twenty-three, making him one of the youngest recipients ever.
People familiar with the first awarding of the prize say it had to do with
developing an improved weapons delivery system for Israel’s fleet of
fighter jets. The second me was in 2001. This award is even more clouded
in secrecy than the first, but it reportedly had to do with his work on a
major project involving the concept of figh ng future wars. One official at
Israel’s Ministry of Defense (who could not be named for security reasons)
said, “What General Ben-Israel developed is s ll one of the main secrets
we have in our arsenal.”
He also led the Analysis and Assessment Division of IAF Intelligence. In
this posi on, it was his job to analyze how the enemy was thinking. In
order to do that, he tried to think like them, pu ng himself in the shoes of
Syrian, Egyp an, Jordanian, Iraqi, Lebanese (and later, Iranian) leadership.
He used these lessons o en when heading MAFAT. During that period he
was also in charge of the Talpiot program. General Ben-Israel would o en
refer to his work in intelligence in his bid to get Talpiot cadets to ou hink
their foreign enemies outside as well as their peers in other army units.
Course work, lesson planning, special lectures and army service were
redesigned with this goal in mind.
In order to accomplish as much as Yitzhak Ben-Israel has, you need to
think a li le differently from your peer group, and much differently from
the average ci zen of the world. And General Ben-Israel has worked
relessly to help Talpiot cadets see into a corner of his mind so they can
get exposure to the kind of thinking that has propelled the Israeli Defense
Forces so far ahead of the na on’s enemies in technology of warfare.
Every few years, General Ben-Israel takes me away from teaching at Tel
Aviv University (where he heads the Tel Aviv University Workshop for
Science and Technology Security) to lead current and former Talpiots on a
week-long crea vity sabba cal, where they can bond, share stories and
informa on, and sharpen their crea ve skills.
During one of these sabba cals, the group of young men and women
from several Talpiot classes (some s ll ac ve, some in reserves) were
driven into the Negev and dropped off at an old air force camp. They were
split into groups and told to come up with an idea of something they’d like
to create, something that can be done in a week. The needed materials
would be provided.
One team came up with a car deodorant that has the smell of a new car.
The team was able to analyze the chemical compounds of that “new car
smell” and reproduce it in a spray. Another team came up with a way to
engineer a small box that could be dropped into the water tank of a toilet
to cancel the sound of the flush. General Ben-Israel’s hulking figure
watched over the demonstra on of this inven on. He asked the young
Talpiots why they’d spend their me working on something like that, and
they replied that some mes they phone their friends while in the
bathroom – but it doesn’t have to be obvious!
The general is gra fied when the young men and women he’s working
with approach problems from an en rely new angle. He says, “We do a lot
of ac vi es like this to help encourage people in Talpiot to think in a
crea ve, out-of-the-box way.”
General Ben-Israel spent his career in the military thinking out of the
box. A er the Yom Kippur War he helped reanalyze crucial data. He s ll
uses the lessons he learned to teach and demonstrate his different way of
thinking.
For instance, before the Yom Kippur War, the intelligence community
looked primarily for evidence to support their conjecture that the Syrian
and Egyp an armies were merely carrying out exercise maneuvers – which
was precisely what the Arab military wanted them to think. Ben-Israel
recalls, “Yet, from me to me there were pieces of informa on that
refuted this conjecture. For example, a few days before the war, the Soviet
Union put the families of Soviet advisors in Egypt and Syria on a special
chartered airplane and flew them back to Moscow. You don’t do this if the
army is merely carrying out an exercise. But the chief of Israeli intelligence
said, ‘Okay, I have so much informa on which supports and corroborates
the exercise conjecture and I have only a few that negate it; therefore I
think the most probable one is the exercise conjecture.’ My method is not
to look for suppor ng evidence. I look for refu ng evidence. If he had
adopted my model, he would have seen two possible conjectures here,
exercise or war. We had strong refu ng evidence against the exercise
conjecture, the Russian families leaving the possible theater of war en
masse. This should have rung serious alarm bells.
“In 1973, if the intelligence community, and especially the Mossad, had
thought about refu ng evidence more, they would have simply followed
up to see that the Arab armies had indeed carried out the exercises they’d
supposedly been assigned. They would have immediately found out that
neither Egypt nor Syria ever actually completed the exercises. It was all
part of a misinforma on campaign. The Egyp ans would send telegrams
saying this unit should do that exercise, this unit should do that, knowing
we’d intercept their communica ons. But we never checked to see that the
armies were, in fact, ignoring the telegrams. That was a fatal mistake.
“In addi on to giving credence to the conjecture that they were
preparing for war, there easily could have been a third op on: there might
be figh ng of some sort, but it might not be a full-scale war. No one
thought of that either.” Ben-Israel used his “refu ng evidence model” and
was proven correct. He concluded, “Same facts; different ways of looking at
them.”
Though former intelligence officers o en get behind the mes, General
Ben-Israel’s legendary, unique way of thinking brings Israel’s intelligence
agencies back to him, unofficially, even today. In late 2011, he had been
analyzing “the Arab Spring.”
“We collect a lot of data on what’s happening around us. Some mes we
know facts, some mes we think we know, then find different opinions.
Some mes we have what we believe are facts that later turn out to be not
true. What is the rela on between what you know, or think you know, and
the decisions that you have to make?
“To make a decision, you have to es mate what will come out of that
decision. For instance, take the Arab Spring. You read a lot about it. You see
stories on television. You send your agents to these countries that are in
the midst of revolu on or to countries that could be the next to see
revolu on. You monitor it. You gather a lot of informa on. But what should
we do about it?
“To answer that ques on you have to assess what will possibly happen
with this Arab Spring. Will you get democracy? Perhaps you’ll get a take-
over by the Muslim Brotherhood. Will these countries go back to where
they were a er a year or two of revolu on? You need to weigh all of the
possible ways the situa on could develop. Some people believe that if you
study it long enough you may calculate what will happen, but I don’t think
so. There is no way on earth to judge what will happen in the future. It is a
logical problem.”
“Here is a simple example. You see one white swan, then a second, third,
fourth, fi h and sixth. You s ll can’t conclude that all swans are white. It’s
impossible, logically. Once you realize it is illogical, what will you do?
Psychologically we were built to trust past experience in similar situa ons,
but if we can’t do that, how should we handle ourselves in the world?
“I think there is a way, though it is not in our nature. Nature built us to
be induc ve, to make generaliza ons from past experience. If you put your
hand in a fire and feel the heat, you will never put your hand in a fire again.
But perhaps the pain you felt in your hand might not be caused by the fire.
If you were a scien st, you might check your facts by pu ng the hand in
from a different angle, or pu ng the other hand in the fire. You’d do all
sorts of tests. This standard way of scien fic thinking can be limi ng and
destruc ve – even deadly – in the world of intelligence. For Talpiot recruits
going into research and development or into intelligence units, as many
have done, it’s crucial to drop that way of thinking.”
Thinking out of the box does not preclude learning from past mistakes,
and this quality is a badge of honor to General Ben-Israel. During his me
as head of MAFAT, and as the de facto head of Talpiot, he really pushed this
point. To this day, when he lectures to a Talpiot class, he tells stories of past
mistakes and tries to get the cadets to understand his way of thinking, so
that they can learn to think differently from the way that most of us are
preprogrammed by nature.
That’s a key training tool not just for students going through a program,
but for young men and women going through life.
Amir Schlachet, who moved on from Talpiot to major posi ons in the
banking world uses Ben-Israel’s kind of thinking to solve problems in his
work. Though it may not be common for most people, Schlachat suspects
that there’s usually something inside a Talpiot that gives him the ability to
think differently. “One of the strengths of Talpiot recruits is that they are
mul disciplinary by nature – oh, and we’re con nuously curious.” And as
any teacher will tell you, the best student is o en a curious student.
CHAPTER 8
REALITY CHECK

Isystem
f academics are the backbone of Talpiot training, the central nervous
is the experience the cadets get going from unit to unit. In each
unit-to-unit visit, the goal is to take the theore cal lessons students
learned back in the classroom at Hebrew University and apply them to real
field exercises and ba le situa ons.
When the Talpiot program started, the military was adamant on one
point: the students had to be a part of the IDF. Its early leaders knew that if
the program were to be a success, the young cadets would have to see the
problems faced by their brothers and sisters in the rest of the IDF so that
they could come up with crea ve ways to help them and to advance the
way the IDF works and fights, on all levels.
Students who had made it into the program soon began visi ng the
different units of the army, navy and air force to give them a feel for what
happens in the field beyond the classroom. But because Talpiot was s ll a
secret and because of rivalries and suspicion throughout the various ranks,
it was difficult to successfully integrate Talpiot students into the rest of the
army.
Ini ally, the unit-to-unit visits were haphazard and disorganized. As
there hadn’t been much me from the moment Talpiot was approved to
the actual start of the program, there was li le chance to get the word out
to field commanders whose help would be crucial. There would be a lot of
“hurry up and wait” – boarding buses, then wai ng outside checkpoints.
Wherever the Talpiot students arrived, accommoda ons and meals were
arranged at the last second. Even acquiring ammuni on for various Talpiot
training sessions was difficult. Though there were excep ons, the students
were usually not made to feel par cularly welcome.
The navy intelligence and technology units, however, were very
different. Officers there were o en more broadly educated. And though
the navy was important, it was not seen as a unit with the history and glory
of the tank units, paratroopers and the air force. Many of the early Talpiot
recruits were given high level access to problems faced by Israeli naval
officers and they were granted green lights to try to help. Several Talpiot
recruits from the first three classes later performed extensive service in the
navy.
Ophir Shoham was recruited to Talpiot’s second class in 1980. He was a
rough and tumble, serious student and soldier. A favorite among his Talpiot
peers, they credit Shoham with bringing their group together.
He quickly became a legend during basic training with his fellow Talpiot
classmates. A young recruit from a paratrooper unit was giving a Talpiot
student a tough me. A er a few minutes, the much shorter Shoham
walked up to the eighteen-year-old bully and told the guy “go shove off –
find someone else to pick on.” The recruit refused and pushed him. But
Shoham had already excelled in extensive premilitary mar al arts training
and the next thing everyone knew, the bully went flying through the air. He
wound up with a broken leg and was subsequently dismissed from the
paratroopers because he could not complete the training.
In fact, in those early days, many in Talpiot felt bullied and abused not
just by other soldiers, but by field training commanders as well. Later, a
special inves ga ve panel sent by the Ministry of Defense would agree.
As Talpiot matured, a war was raging to Israel’s north in Lebanon. Years
of terrorist a acks on Israelis directed by the Pales ne Libera on
Organiza on and other terrorist groups angered the Israeli government,
led at the me by Prime Minister Menachem Begin. (These groups had
virtually taken over Beirut and other parts of Lebanon during a power grab
in Lebanon’s civil war, which began in 1975.)
Two a acks in par cular were so outrageous the Israeli government felt
there was no choice but to respond. First there was the hijacking of a bus
by terrorists who had infiltrated northern Israel through the Lebanese
border. That a ack resulted in the murder of thirty-eight Israeli civilians,
including thirteen children. Then there was the a empted assassina on of
Israel’s ambassador in London, Shlomo Argov. The shoo ng le the
ambassador in a coma for months. Though he survived, he was paralyzed
and later lost his eyesight.
The war in Lebanon quickly became controversial in Israel, as many
ci zens felt Israel had entered into a full-scale war that had been
avoidable. Many argued that there were other ways of defea ng the wave
of terror cascading from Lebanon.
For be er or worse, war o en leads to innova on, and in this case bright
Talpiot students and Israel’s military research and development arm were
there to take advantage of the opportunity the war brought them. Lebanon
became a military training ground for some early members of Talpiot. They
were able to witness war from the frontlines – seeing which weapons
worked, which caused problems, which systems needed remodeling. The
Israeli military machine needed to come up with new ways of figh ng, and
Talpiot was there to help.
One system born in the war would later become known as “The Trophy.”
It is a tank-mounted an -rocket device that is now in service on Israel’s
fleet of Merkava tanks. It automa cally fires a projec le at an incoming
an -tank rocket to disable and misdirect the projec le, saving the lives of
the tank crew inside. Talpiot students and a few of their instructors would
become instrumental in making “The Trophy” opera onal.
As the controversial war se led into a controversial Israeli occupa on of
southern Lebanon, Talpiot’s unit-to-unit program began to take form in
more concrete and ins tu onalized ways. The army, air force and naval
units started to come to the understanding that the Talpiot program was
here to stay, that it was not a unit for spoiled nerds and that the students
of Talpiot were indeed there to help them and future IDF fighters. They
became more and more forthright in explaining their problems, their
successes and their failures – and most of all, outlining how and where
they needed help.
Coopera on and special training for Talpiot cadets became more
encouraged and more accepted. Eventually, a very specific and organized
program was devised to help give Talpiot cadets a real taste of what life
was like in the skies over Israel in warplanes, in the hangars where the
planes were fueled and repaired, in the trenches with the “green army,” in
tanks, armored personnel carriers, in the troop carriers that ferry
paratroopers to their drop-zones, on ships at sea and even some mes in
actual warzones.
The training sessions with figh ng units started to build connec ons
between Talpiot and those units. Talpiot cadets would actually do the work
of their fellow soldiers in the field. They didn’t just learn about changing
re treads on tanks, they changed them. They didn’t just learn about tank-
mounted weaponry, they actually got into the tanks, drove them through
obstacles, iden fied targets and fired. They flew in fighter jet simulators,
they fired machine guns and ar llery, they dropped off explosives with IDF
engineer teams and they jumped out of airplanes with the paratroopers.
They sailed on the sea with naval commanders and went underwater in
Israel’s submarines.
Soon the ordinary soldiers and officers in those units started to do
review training with their Talpiot colleagues: they wanted more me to
explain their challenges in the hope that Talpiot research would help them
and their units down the line.
Ophir Kra-Oz spent some me in the United States while his father was
an execu ve working in Georgia. He was inducted into Talpiot’s thirteenth
class in 1991, soon a er Iraq a acked Israel with volleys of SCUD missiles.
He believes that going from unit to unit helps Talpiot cadets draw closer
to the rest of the Israeli Defense Forces. “A er you li a 45 kg shell and you
smash it on your hands and move it around, you can say ‘okay, that’s very
heavy,’” he says. “Then you say ‘let’s try to make them lighter, but with the
same power.’ It’s beneficial to see the problems on the ground. Though
you’re only eighteen, you see so many parts of the military you’re really
ge ng more informa on than most generals: they only get to see what’s
under their command. Most generals may be an expert on one thing and
have their own department, but they never got this perspec ve of the
army.”
Spending me in so many units later allows a Talpiot graduate to think
more seamlessly, Kra-Oz notes. “Almost all of the projects are integrated
and that’s a great advantage. Compare this to someone who studied
something at university and went on to a company with a very specific
focus. They get more and more specific as their career goes on. Although
we [Talpiot cadets] come from math, physics and computer science
backgrounds, we need to see how to implement the theore cal into real
systems and real products which help the end user, in this case the soldier,
but in an integrated way.
“For instance, take Iron Dome, an an -missile system that targets short-
range and medium-sized rockets. It was developed in part by Talpiot
graduates from an idea of a Talpiot cadet. You need new technology and
you need it really fast. It’s a mix of at least five to ten different
technologies. You have the missiles and the ballis cs, but you need the
so ware to know that this missile isn’t going to hit a populated area.
Seeing as much as we saw in the army really helped us understand how
everything comes together and how everything is connected.”
Saar Cohen, a graduate of Talpiot’s fi eenth class in 1996, agrees that it
was instruc ve to see the diverse needs of the military machine and the
different technologies that keep it moving. “That was what I was most
interested in from the start. I wanted to hear about military technology –
and they were pre y open about it, with the hope that we would have
ideas to make it be er, more sophis cated, more protec ve and easier to
use. By the me I was in Talpiot, they encouraged us to do a lot.”
Cohen was also drawn to the camaraderie in the different units. “All
units have their own sort of code, and this was fascina ng to me. And
more importantly, it makes you directly think about why you are working
so hard. There are real people out there; and to some extent, just as we’re
depending on them to hit their targets, they’re depending on us to break
enemy code, to come up with be er intelligence, to reconfigure weapons
and to use physics to give them a greater edge in the field. It really hits you
when you’re out there. You’re exposed to life and death situa ons and
military secrets at a very young age. It’s sobering to see things at that level
as an eighteen-or nineteen-year-old. It quickly makes you realize just how
important a part of the machine you really are.”
Teamwork was another lesson Cohen learned from the army. About
working on so ware programs while in the military he comments, “In most
cases, you work in a team. There is nobody looking directly over your
shoulder when wri ng code. But there are managers and supervisors, like
anywhere else. Despite less supervision than in the corporate world, you
have a serious amount of pressure in this kind of environment. But you
always have someone to seek out and talk to for advice. Most importantly,
people are always helpful on your team. It’s a mission. Everything you do,
every program you work on, is a mission and we all view it that way.”
Today the length of the unit-to-unit training sessions varies from two
days to two weeks, depending on the branch, the complexity of the unit’s
work, and how much help the unit might need in the future, according to
research and development planners at the Ministry of Defense. Almost
every sec on of the Israel Defense Forces will host Talpiot students in the
weeks when they’re not studying. There is no summer break for Talpiot.
Most of the cadets will tell you that going from unit to unit is one of their
favorite parts of the en re Talpiot experience.
CHAPTER 9
ATTACK BY KEYBOARD

Ionn northern
July 2013, amid the Syrian civil war – while shells periodically showered
Israel, while Iran con nued its produc on of nuclear material,
while the poli cal unrest in Egypt allowed Hamas to arm itself – Israel’s top
two military leaders found their way to a nondescript building hidden by
trees in central Israel. The only armed soldiers in the area were the guards
at the gates and doors of the complex.
For the first me in Israel’s history, Chief of Staff General Benny Gantz
(Israel’s highest ranking officer) and Minister of Defense Moshe “Bogie”
Ya’alon (a former chief of staff) a ended a special ceremony for the men
and women in a unit known inside Israel as “8200.” In most armies, it isn’t
an everyday event for the top brass to visit a facility where minds are more
important than brawn, where keystrokes are as important as firing
weapons on the ba lefield.
Unit 8200 is a fairly large, but elite, group of soldiers who work on
computers all day. They can hack into just about any military network in
the world. It is rumored that Unit 8200 can tap into electronic systems of
enemies far and near, turn off power plants, radar sta ons and the
electronic capabili es of enemies and allies alike. Unit 8200 has become
just as important to Israel as the men in tanks and the pilots who fly F16s.
One source familiar with Israeli military opera ons said, “8200 is now
involved in just about everything we do.”
The exact reasons for General Gantz and Ya’alon’s congratula ons to
8200 remain classified, but it’s clear the unit had done something
par cularly significant. When General Gantz addressed the unit, he
focused specifically on its covert role in intelligence: “Intelligence
transmi ed in real me enables the IDF to create a clear and accurate
picture at all mes and gives impetus…for sharp and fast ac on, which
proves powerful on the ba lefield.” In his remarks, the Defense Minister
added, “Your ability to iden fy threats in a mely manner leads to
preven on. This unit is an example of the proper way to deal with frequent
changes in the technological world around us. New threats create new
arenas.”
While scores of Israel’s top high school computer students are recruited
for 8200 each year, it is Talpiot graduates who play an outsized role in
commanding and crea ng programs for this unit.
Among the responsibili es of Unit 8200 is the opera on of a massive
listening and signal intelligence-gathering facility capable of intercep ng
informa on all over the world. While 8200’s capabili es are global, one of
its main responsibili es includes listening in on what is happening not far
from Israel’s borders, in Gaza and inside the disputed territories in the
West Bank. The unit has been credited with foiling scores of terrorist
a acks and for helping Israeli security forces make preemp ve arrests.
Reports s ll unconfirmed by Israel say that in September 2007 eight
Israeli jets took off from a Negev air base. Their mission: destroy a Syrian
nuclear reactor that was under construc on in the eastern part of the
country, not far from the Iraqi border. The jets were able to straddle the
borders of several countries, including Turkey, in order to confuse radar
systems. Reports from outside of Israel say that Unit 8200 also played a
role by breaking into Syria’s radar defenses and limi ng its ability to see the
incoming Israeli planes. The jets successfully fired their missiles and
dropped their bombs before safely returning to base in Israel.
Soon a er, reports from outside of Israel gave programmers at 8200
credit for scoring another major victory, this me against Iran’s nuclear
program. They had been asked by the prime minister’s office, reportedly in
coopera on with the Mossad, to develop some sort of a virus designed to
infect, disrupt and spy on computer work sta ons in Iran that were being
used to work on the Islamic Republic’s nuclear program. Their answer:
Stuxnet.
Stuxnet is a computer worm that was used to infect computers in Iran
and was also used to give outsiders control of Iran’s centrifuges (or at least
to cause Iran to lose control of those centrifuges) as they purified nuclear
material to a level used in bombs and or missiles. The United States also
reportedly played a major role in Stuxnet; and it is congruent with US
strategy to disrupt and delay Iran’s nuclear ambi ons without physically
a acking nuclear plants.
Former Mossad chief Meir Dagan was asked about Stuxnet during a 60
Minutes interview in 2012. It would have been treasonous for him to
formally comment about Israel’s role in such an ac on, but he smiled
widely – promp ng many to feel that was confirma on enough.
Since nobody has claimed credit, it’s impossible to judge whether
Stuxnet was truly a success. While it did at least slow down Iran’s nuclear
program, the program may have been designed to inflict more damage or
to spy on the program over a longer term. So its overall success has to be
ques oned.
Stuxnet wasn’t the only a ack launched on computers inside of Iran’s
nuclear facili es. A virus called ACDC struck Iran’s Natanz and Fordo
nuclear facili es in the spring of 2012. One vic m inside Iran with
knowledge of the malware was quoted on an Internet message board
saying, “There was also some music playing randomly on several of the
worksta ons during the middle of the night with the volume maxed out. I
believe it was playing ‘Thunderstruck’ by AC/DC.” A UPI story on the
incident said the message could not be verified, but it was believed to have
come from workers at the Atomic Energy Organiza on of Iran.
Ophir Kra-Oz, whom we first met in chapter 5, had already moved on
from the army well before Iran’s computers were infected with this high-
end spyware. But well before the Stuxnet computer a ack program was
launched, Kra-Oz, of Talpiot’s thirteenth class, played a big role in Unit
8200. While in Talpiot, Kra-Oz was eager to learn everything about the IDF
that he could. He took full advantage of Talpiot’s unit-to-unit field trips,
learning about the ar llery, the armored corps, the navy, Israel’s space
agency, combat troops, fighter jets, radar and weapons-firing systems.
But his heart was always in technology. A er gradua ng from Talpiot’s
academic program, Kra-Oz moved on to 8200, working on developing
so ware that retrieved data stored in the Israeli military machines’
computer servers. He described it as a “google-type search engine” for the
army. It mined vast amounts of informa on for intelligence agencies and
other branches of the military with specially cra ed algorithms. The system
could find very specific informa on very quickly and it was designed so
that many people with many backgrounds could quickly understand how to
use it.
Kra-Oz started as a programmer in 8200, then became a team leader and
later one of the youngest sec on heads in the elite unit’s celebrated
history. Later in his career, in discussing his experience with venture
capitalists, Kra-Oz pointed out that 8200 was a large unit, but it operated
like a series of small start-ups, with different teams working quickly on
different projects, although always in touch and in coordina on with each
other.
While the rigorous Talpiot program helped prepare him for 8200, the
pressure was s ll intense. “In 8200 the military is physically very close to
you, is very demanding and always has strong opinions about a project. We
might be given days to do what we’d be given a year to do in civilian life. If
we lost informa on, it could be cri cal; someone’s life could be in danger.
If you know that this guy is carrying a Qassam rocket and is on his way to
fire, that’s pre y serious. And I was able to help come up with programs
that allowed the military to defend against those kinds of threats.
“My me in the army also taught me a valuable lesson I’d need later in
civilian life: you have to delegate. There’s no way to sa sfy a military client
because the problem is endless. You’re constantly gathering informa on
and insights from all over the world in many different languages, in real
me, with limited resources. On the other hand, they had no financial
string to pull as there is in the corporate world. Payments can be denied in
civilian life. In the army the worst they could do was yell at me and tell me
they were not happy.”
A good number of other Talpiot graduates par cipated in vital
intelligence ac vi es as well. Born in Argen na, Adam Kariv o en felt like
an outsider a er his family moved to Israel. Lacking many of the
connec ons and networks other families had, he was convinced that he
would never be able to be successful, as Israel seemed to be a place where
connec ons are vital. When he saw a TV report about Talpiot, it clicked.
This is where he wanted to go. Adam didn’t think he’d get in, but he got
through the difficult tests and became part of the eighteenth class of
Talpiot in 1997.
A er gradua ng, he was sent immediately to a technology unit of the
Israel Intelligence Corps. He’d spend the next nine years working as a
so ware engineer and then a unit leader. Everything he worked on remains
highly classified, but it was his job to come up with new ways for the army
to monitor events going on around Israel and to track people Israel needed
to watch – from high-ranking officials to terrorists possibly planning
a acks.
The pay he received came to about four hundred shekels a month; that’s
about 125 dollars. “Compare that to the thirty or forty thousand shekels a
month you’d get in the private sector,” Kariv laughs. “But you do the work
to the best of your ability because if you don’t get something done it could
make a big difference in the life of someone, maybe a fellow soldier on the
front line trying to protect you and your family. That’s a very big
responsibility and every day I did everything I could to do my best. Every
keystroke meant something. It takes a while before it sinks in that your
work could mean life or death. And even if it isn’t life and death every
minute, even in those mes when it is less immediate and less cri cal, it is
s ll very important.”
Another Talpiot dra ed into military intelligence was Haggai Scolnicov. “I
went to a branch that is top heavy with great math and science people,” he
recalls. “It is a small place where mathema cians are needed. It was a
shock to get there. There’s a lot to learn and absorb. It’s an amazing group
of people. I worked on data analysis and algorithms. My unit had a very
ght and specific domain which had been developed for several years. In
certain parts of Israeli intelligence it is very well known, but it is not the
kind of thing that gets much outside a en on, and that’s by design. Our
base is north of Tel Aviv and it houses many similar units that have almost
no connec on to one another as we’re not always supposed to know what
the other units are doing.”
Scolnicov is very proud of what he did in that unit but he can’t share
everything saying, “We’re all quite good at talking about what we did
without telling you anything we can’t. We know how to work around the
details.” He con nues, “No two projects were the same. It’s very exci ng.
It’s not like working for some client somewhere; you’re doing something
for your country. And you have an immediate, some mes very clear impact
on security and in the success of Israel’s military. They threw a lot of
important things at us. I was involved for ten years with problems that
were generally considered unsolvable. We solved a good number of
unsolvable things. Remember, in Talpiot they teach that nothing is
impossible. They didn’t quite mean it because, of course, some things are
impossible. But they trained us to think that way. If you think differently
about the ques on or find the loophole, you can move the unsolvable
forward.”
Uri Barkai also did his post-Talpiot service in an elite intelligence unit
that served as a main nerve center for the IDF. He came into the program
as a so ware expert. Although his role was very important to Israel’s
intelligence-gathering machine, he was never given the full picture of what
his work was used for. It remains classified.
Barak Peleg (of the twenty-first Talpiot class of 1999) went into signal
processing. He was tasked with developing so ware to track radar, radio,
computer and other communica on footprints le behind by people of
deep interest to the IDF. That includes armies throughout the Middle East
and terrorist organiza ons including Hamas, Islamic Jihad and Hezbollah.
All of them are becoming more sophis cated in how they use electronics
and how they communicate with each other and with the governments
that help them financially and with training. The list is long and includes
Iran, Syria, Lebanon and many other Middle Eastern countries with close
es to terrorism.
He explains: “Signal processing would be obtaining the signal, digi zing
it, manipula ng it to handle whatever happened to it. If it’s a transmission,
how the medium affected it, and cope with what the medium did to it. This
is for incoming and outgoing signals. The data would then be analyzed and
then analyzed more in depth by army intelligence, AMAN. It could really
give them a window into what was happening in places well beyond our
borders.”
Looking back on his Talpiot experience, Peleg conjectures, “The most
important thing about Talpiot – which many graduates don’t think about
and the public doesn’t know – is that they make us fearless. It becomes
very hard to scare you. You can tackle anything.”
In 2010 the task of tackling the unknown was given to General Yitzhak
Ben-Israel, who formerly had been head of MAFAT and a Talpiot lecturer.
By then, it was crystal clear that ba les were being fought between
countries online and that online supremacy would be a key ingredient to
any future military campaign. China and the US were already figh ng
ba les online. China had been accused of breaking into the computer
networks of American companies and stealing informa on, even snooping
on highly classified military informa on.
Iran was also becoming more and more adept at cyber-warfare and
espionage through computers. Computer hackers in the Islamic Republic
broke into Saudi Aramco’s computer system, wiping out key informa on.
Computer users in Iran have also been accused of a acking the financial
system in the United States and crashing or slowing the websites of
American banks. Countries must be able to protect their financial and
physical infrastructure from enemies using computers thousands of miles
away.
General Ben-Israel was appointed cyber adviser to Prime Minister
Benjamin Netanyahu. One of Ben-Israel’s first moves was to create the
Israel Na onal Cyber Bureau on August 7, 2011, and appoint Talpiot
graduate Eviatar Matania as its head. The goal of the INCB is to provide the
prime minister with advice on managing this new and crucial front, where
both defense and offense are needed and to carry out missions. It is also
expected to provide for the con nua on of “life as normal” if the country
comes under some sort of cyber-a ack, just like the Home Front Command
is expected to do if and when Israel comes under physical assault.
Another reason for the crea on of the INCB was to expand Israel’s lead
over its enemies in the Middle East in the field of cyber-warfare. As Israel’s
enemies grow their own cyber capabili es, the INCB was created to
maintain that qualita ve edge so cri cal to Israel’s survival.
Matania was invited to a mee ng of the Israeli cabinet in November
2011, shortly a er he became head of the Israel Na on Cyber Bureau. He
told members of the government that cyber a acks are “a broad threat to
human society. While this is a challenge to the state, it is also an economic
opportunity. The more we invest in academia and industry, the greater the
return we will receive, from both economic and security perspec ves.”
Prime Minister Netanyahu followed Matania, explaining to the Cabinet,
“Israel is a significant force in cyberspace.… Just as we developed the
unprecedented Iron Dome system that successfully intercepts missiles, we
are developing a kind of ‘digital Iron Dome’ in order to defend the country
against a acks on our computer systems. The INCB is designed – first and
foremost – to organize defensive capabili es based on coopera on
between three elements: security capability, the business community and
the academic world.”
In 2012 Matania built a na onal cyber situa on room to assess threats
launched against Israel from foreign computers. Its goal is to have one
central place where Israel’s poli cal leaders can go to see the full picture of
what is threatening the state and what’s being done to protect it. It is also
a place where high-ranking military officers, government officials and
business leaders can come to share informa on.
The Israel Na onal Cyber Bureau works closely with Israeli so ware
companies to protect the na on from the growing threat of hackers
working for hos le governments, terrorist groups, or working as lone
wolves on the Internet.
Another of Matania’s ini al goals was to create clear and direct links
between the bureau and computer scien sts working in Israeli industry
and at top Israeli universi es including Hebrew University, Tel Aviv
University and the Technion. This mul -system, mul -organiza on kind of
project management is an approach Matania developed in Talpiot, where
sharing informa on and coopera on are highly prized. He also ins tuted
awarding funds to promising minds and programs in the cyber field. In
2013, $20 million was set aside for individuals, companies and universi es
with good ideas in the field of cyber-security.
Eviatar Matania has wisely used the bureau to help adver se Israel’s
prowess in global cyber-security, crea ng thousands of jobs and billions of
shekels in revenue. It also serves as an arm that cooperates with friendly
foreign countries and shares informa on about threats and enemies, much
like Israel’s intelligence agencies. The INCB also serves as a gateway for
foreign investment in Israeli’s technology sector.
In late 2012, the INCB took a new step: establishing a research and
development arm. It was similar to the step the Ministry of Defense took
decades before, inves ng heavily in research and development for Israeli
made weapons. The research and development arm of the INCB is known
by the acronym MASAD. It deals with cyber projects for both the military
and private sectors. Israeli start-up companies, programmers in established
so ware companies, university professors, members of the government
and the defense establishment have all been asked to contribute to the
effort.
As MASAD was announced, MAFAT director Ophir Shoham (another
Talpiot graduate) issued a statement saying, “The plan is an addi onal layer
in the Ministry of Defense’s prepara ons to meet the cyber challenges
currently facing the State of Israel. The MASAD plan is expected to link
technological vectors based on the know-how and capabili es of
companies and academia with common defense and civilian needs.”
Today, through INCB and MASAD, Israel is always at the cu ng edge of
cyber development. Warfare has changed considerably since the 1948 War
of Independence, when a clumsy, inaccurate canon known as the Davidka
could turn the de in ba les simply by scaring the enemy with its piercing
shriek and massive explosive boom. Its clever inventor discerned that
when an army is outnumbered, ingenuity could compensate for lesser
troops. In that respect, the crea ve and searching minds of Unit 8200 and
the INCB con nue to uphold Israel’s resourceful military legacy.
CHAPTER 10
MAKING AN IMPACT

Icredit
n the Israel Defense Forces, pilots and paratroopers get most of the
for defending the na on. Tanks make up the backbone of the IDF. A
highly visible Israeli navy patrols the coast on the Red Sea, and from Gaza
up to the Lebanese border on the Mediterranean coastline. When it comes
to intelligence, the Mossad is feared abroad and venerated at home. When
violence breaks out, the cameras go where the ac on is. Reporters speak
with the crews inside Iron Dome missile defense ba eries. They film
soldiers of the Giva and Golani brigades with machine guns slung over
their soldiers, carrying heavy backpacks and ammuni on. The media
records the voices of helicopter and F16 pilots describing their missions
while hiding their faces to protect their iden es.
Talpiot’s members are rarely interviewed on domes c or foreign news,
and the program’s massive contribu on to all of those defenders of Israel is
invisible to the public. Yet it’s really the Talpiot graduates who spent so
many years consistently helping to make the IDF and its impressive arsenal
so effec ve during war me and during the quiet mes between conflicts.
In fact, Talpiot graduates have come up with so many ideas, designs and
updates to Israel’s weapons and technology arsenal, an official count is not
even kept by the Ministry of Defense.
The contribu on of Talpiot to Israel’s defense went far beyond anyone’s
imagina on, primarily in three areas: research and development; the
Israeli space program; and electronic warfare.
Israel’s space program grew up alongside Talpiot. Its graduates have
contributed directly to Israel’s space program by crea ng space vehicles,
electrical systems, communica ons systems and by working on the
cameras carried by Israel’s satellites.
In Israel’s mind, staying ahead of the Arab armies and navies isn’t
enough. It also must stay ahead of the more advanced technologies from
first world na ons supplying weapons and weapons systems to the na ons
that pose a constant threat to Israel.
In the early stages, when the Talpiot project was first ge ng started, the
navy benefi ed most from its efforts, largely because it was more
welcoming to Talpiot than other branches of the military. Just prior to the
forma on of Talpiot, the navy had done some impressive soul-searching.
It had suffered a terrible loss four months a er the huge success of the
Six-Day War: a brand new breed of Russian missile, fired by Egypt,
slammed into the Israeli Naval Ship Eilat while it patrolled interna onal
waters in the Mediterranean Sea. While the crew of the Eilat waited for
evacua on and rescue from other Israeli ships, Egypt again fired on the
wounded vessel, sinking her. Forty-seven Israeli sailors were killed in the
a ack, another forty-one injured. Israel was shocked into becoming much
more serious about defending its fleet at sea.
The six years between that deadly a ack and the 1973 Yom Kippur War
was very busy for Israel’s navy. The sea force drilled. They updated their
equipment. Officers studied the naval successes of allies abroad.
And it all paid off. The Israeli navy was one of the few branches of the
IDF to perform excep onally well during the Yom Kippur War. Though the
general staff was star ng to realize how valuable the navy could be, only a
small part of the defense budget was earmarked for the navy.
When the first classes of Talpiot students graduated their Hebrew
University course work many began to gravitate toward the navy. Eli Mintz
had become an expert on data mining during his post-Talpiot military
service. He was a pioneer in developing programs for the Israeli navy using
algorithms to improve radar systems. Mintz says, “The Israeli navy was very
small in the 1980s, but it was quite sophis cated. I was very mo vated in
the navy to learn and apply what I learned.” To Mintz, it was a two-way
street. He wanted to help the navy on its course to innova on, but he also
wanted to learn what the people who were already there were working on
and how they were developing state of the art hardware and so ware.
“One of the best things about Talpiot,” Mintz recalls, “was that a er
gradua ng, you had your choice of pos ngs, assuming the place where you
wanted to go would take you. Nobody else in the army gets to do that. So I
picked a project in the navy where I did a lot of project management. It
started with algorithms, but it evolved into managing certain aspects of a
project from the technical side. I was able to wear two hats, doing real
technical work and also project management.”
The exact project Mintz worked on remains classified. “You don’t
develop i-Pods in the IDF. You develop weapons. We developed a weapon,
which has since been improved by others in research in development.” The
weapon he worked on has been deployed, but it has not yet been used
because Israel s ll hasn’t fought the kind of extensive naval ba le where
such a weapon would be used. “If the weapon is used, it will have a big and
immediate impact,” he adds confidently.
Gilad Lederer is one of the most interes ng and colorful of the Talpiot
graduates. Lederer also joined the navy and became one of Talpiot’s first
combat officers serving on a missile ship. (His post-army work took him to
Africa, including many countries in the midst of civil wars. More on his
amazing post-army business journeys in a later chapter.)
Growing up in the 1970s, Gilad learned to sail as a boy and he always
loved the sea – though that was not common for the average Israeli boy.
A er Talpiot gradua on, his first assignment was to the naval academy. He
served on a Sa’ar 4, a fast missile boat (about 400 tons and 190 feet long).
He rose in the ranks to become a bridge commander before moving back
into research and development. His opera onal knowledge learned at sea
combined with his Talpiot training made him extremely valuable to Israel’s
navy.
Lederer went to work on developing and improving electronic warfare
systems for the navy. Specifically he worked on “passive electronic warfare,
defense systems designed to monitor the communica ons of other ships.”
Lederer also worked on missile evading electronics designed to help Israel’s
naval ships track and dodge incoming missiles fired from the sea, the shore
or the air. “If you can mess with their homing systems, they can’t hit you,”
he says cheerfully. He also went on to work in ship design, coming up with
ways to make Israel’s ships harder to detect with radar and harder to strike
with missiles.
Ziv Belsky is currently a leader and innovator in Israel’s s ll fast growing
pharmaceu cals and medical devices industry. While doing his service,
however, he was a true innovator. He also was one of the first Talpiot
cadets to become a combat officer. From Talpiot, he also went on to the
naval academy. His assignment was to serve as the execu ve officer of a
Sa’ar 4.5 class missile boat, the most advanced in the Israeli navy. Belsky
brags, “It even had two helicopter pads.”
A er serving at sea, he was transferred back to land at the Israeli navy’s
research and development headquarters. Somehow, during this me he
managed to study electrical engineering and earned a master’s degree. His
job soon became to develop and innovate electronic warfare systems for
use at sea. He and his colleagues came up with new technologies to defeat
missiles fired at Israeli ships using electronic defenses. He says, “If you have
a fast missile and you can shoot it down, great. Rocket to rocket. But if you
have a system that misguides and tricks rockets, that’s be er and more
consistently effec ve. You need tricks.”
For Talpiot graduate Ra’anan Gefen, the navy also was the path to the
Ministry of Defense’s Research and Development branch. A er Talpiot, he
served as a naval officer, then began developing new radar technology and
an -missile systems. To move his ideas forward for the benefit of Israel and
the United States naval forces, he shared Israeli naval technology with US-
based military contractors. About his work Gefen says, “A ship must be able
to defend itself against all threats. Radar is for naviga on, to detect hazards
in the water, to see other ships outside your line of sight. You need surface
radar, to defend against airplanes and drones. Without radar, the ship is
alone and in the dark.”
But it’s not enough to have a good system in place, Gefen insists. The
crew must be able to use it properly. During the Second Lebanon War in
the summer of 2006, Hezbollah launched a shore-to-sea missile (likely
manufactured in China). It slammed into the INS Hanit, which was
patrolling the Mediterranean in interna onal waters adjacent to Beirut.
Four Israeli sailors were killed, but the crew managed to get the ship back
to Israel for repair work.
The problem in this case was simple. The officers on board the Hanit
failed to ac vate the ship’s radar and an -missile capabili es, thinking
Hezbollah lacked the technology to hit the ship – despite warnings from
naval intelligence that Hezbollah did have a shore-to-sea missile capacity.
To this day, Gefen is s ll dismayed over the decision not to use its an -
weapons system to guard against such a acks.
Talpiot has also had a tremendous impact on communica ons in the
Israel Defense Forces. Because Israel is a small geographic area, armed
forces transmissions can easily be picked off by even rudimentary
eavesdropping equipment in Lebanon, Syria, Jordan, Egypt and Saudi
Arabia. Developing ways to encrypt and keep those communica ons secret
is a major priority.
One of the early pioneers in this field is a graduate of the second class of
Talpiot, Boaz Rippin. His army service was dedicated to making radio
signals impossible to intercept by the enemy in the mid-1980s. Israel was
at war during this me in Lebanon – figh ng Yasser Arafat’s PLO, then the
Shiite-backed groups, Amal and Hezbollah.
Rippin had never heard of Talpiot before being chosen for the program
in 1980. He admits that when he joined the unit he wasn’t sure it would
work out. “It was a gamble. Nobody knew what graduates could do with
what they’d learn. I felt like I was part of an experiment. The program was
constantly changing.”
Born in Tel Aviv, Rippin was eleven years old when the Yom Kippur War
broke out. “An alarm sounded in Tel Aviv as rockets were shot at the city.
Car lights would be dimmed. Apartment lights would be shut down so the
city could not be seen by bombers above. I watched a lot of television.
There were reports of people dying. I was worried about my father, he was
a surgeon. He was wai ng at a field hospital near Tel Aviv. I knew I was in
danger. The fear of losing the war was palpable. People spoke about what
would happen if we were conquered and defeated. No one thing shapes
who you are but, the war did shape who I wanted to be in the army, to a
large extent. I wanted to do as much as I could to help. This was part of the
reason I decided to give extra service. I wanted to have an impact.”
David Kutasov was nine when he moved from Lithuania to Israel, and he
didn’t speak a word of Hebrew. He recalls now, “It’s amazing how quickly a
nine year old can learn.”
Kutasov remembers training with paratroopers in Arab parts of the West
Bank and doing a drill where he and the other members of his platoon
were supposed to sneak into the area at night. “The villagers woke up in
the morning and came out of their houses. They spo ed us immediately
and started laughing at us. You have to understand that I grew up in Holon
[just south of Tel Aviv], completely shielded from the territories and the
Arabs. We were taught to ignore Arabs; not to discriminate, but just to
pretend that they don’t exist. Here I was in the West Bank, and all of a
sudden it turned out that not only do Arabs exist, they don’t like us very
much. It was a big shock.”
To this day, he’s credited with changing the way Israel’s ground forces
operate. “The projects I worked on had to do with enhancing the figh ng
capabili es of tanks and infantry, using all sorts of advanced technology. I
haven’t been in the army in twenty years, but the problems I worked to
solve surface today in Lebanon and Gaza. One of the men that was in the
program with me and stayed in the army recently told me that something I
worked on is now viewed as a bible for the people working in the tank
corps – but I don’t think I can say more.”
As Kutasov discovered, for Talpiot grads there are no drills. Everything is
real. You are given technical challenges to write programs or build
something that is cri cal to the security of the state. That adds to the
pressure and the desire to get the job done fast and done well.
Mor Amitai is a legend among legends in Talpiot. Many of his colleagues
and Talpiot comrades say he can literally figure anything out. A er finishing
his Talpiot course work, designing communica ons systems became his
focus.
A member of Amitai’s team on a communica on project for the army
told of an instance when it was crucial for the IDF to know if the answer to
a certain ques on was yes or no. The army would have to do things a very
different way, depending on that answer. They needed to know if
something specific was possible. “If the answer is yes, it is usually easier to
prove. Something exists. If the answer is no, some mes it is harder to
prove. In this case, we were in between. We all worked so much and so
hard on it that we all believed it was impossible. Some mes when you try
very hard at something and fail, it is very close to proving that it is
impossible.”
The project is s ll classified and shedding light on the specifics could
lead to catastrophe for men in the field. The man on Amitai’s team
con nues, “It was a program for a complicated system that should perform
under different condi ons. In the army, you don’t control the environment.
For one soldier in the field, anything can happen, even if he’s trained well.
He can trip, fall, or drop a weapon. This ques on was similar in some
regards. It was a big ques on for the army. Will it perform well enough
under certain extreme condi ons? You cannot test these condi ons unless
you…” He laughed and said, “I really can’t tell you more.” All he could add
was, “The army could not func on well without it and it was something the
army needed all the me. The army uses the system quite widely.”
In his five years of service, Amitai was responsible for complex
components of communica ons systems. Some mes he worked on
projects from scratch. Some mes he had to alter something already in
existence. And some mes he had to combine different kinds of systems. A
lot of the work was analyzing what can go wrong and how it can be
updated in the future, as the specifica ons of different systems need to be
updated for field use.
As in the example above, Amitai’s work in army communica ons always
had to take the unexpected into considera on. A colleague familiar with
his work explains, “It’s like a car. A car is something you can test. It goes
well, air condi oning works fine, everything. But a car should also work
well under circumstances that are not under your control. This is why
manufacturers invest in crash test simulators to see what will happen when
another driver makes a mistake. In the army, you also don’t control the
environment. The enemy is out there; it’s worse than when you drive a car.
Someone’s not making a mistake – someone’s trying to make you fail on
purpose – to kill you and your friends. It’s like building a car to withstand
the tac cs of other drivers who are trying to run you off the road. So when
you design such a car, you don’t have a full picture: you have to think what
the other drivers will do or what the weather can do to you. We spent
most of our me analyzing what we were building, to see if it will survive
awful condi ons.”
Life and death are definitely mo va onal, and might in midate some
individuals. But Mor and his team were constantly told, “You belong to a
small group of very talented problem solvers. The army has invested a lot
in you. Talpiot is the longest course you can take in the army, even longer
than pilot training and service. We’ve invested in you. Now go make us
look good and don’t fail.”
CHAPTER 11
HIGH-TECH TINKERERS

IFollowing
ntelligence is where Talpiot has had one of its most drama c impacts.
the Agranat Commission’s close scru ny of the failures of
intelligence in the Yom Kippur War, The Israel Intelligence Corps was
created. The corps includes the famous Unit 8200 (discussed in chapter 9)
which creates so ware programs, search systems and Internet defense
systems to repel cyber-invaders.
The Intelligence Corps also works on tracking signal-based intelligence
which includes monitoring radio frequencies, tracking telephone calls as
well as other electronic signals. The corps also monitors and analyzes what
is known as Open Source Intelligence. That includes monitoring the media
in foreign countries including newspapers, television sta ons and radio
broadcasts. In many authoritarian countries, the government uses state-
controlled media to control its ci zens, and some mes sends messages to
the West by way of their media.
One of the first intelligence field assignments given to a Talpiot graduate
was in 1982. Opher Kinrot (recruited to Talpiot’s second class in 1980)
found his way into Israel’s burgeoning new intelligence forces. “Israel was
pulling out of the Sinai at the me. Prior to that, when they had bases and
intelligence equipment in the Sinai, they could listen to and watch the
Egyp an army. Now the army needed capabili es to access the same
intelligence, but from farther away. I worked on making that a reality.”
Another game-changing Talpiot graduate has become a modern-day
Israeli Renaissance man. For security reasons his name can’t be published.
He’s a nkerer by nature, and since he was a child he has liked to build
things. A er gradua ng from Talpiot’s coursework at Hebrew University, he
declared his desire to be in “the real green army.” Equipped with small
arms, he would get his chance as a roving tank killer. He became Talpiot’s
first commander in the armored car division, and he set a sparkling
example for others to follow.
For four years, his job was to track enemy tanks with small bands of
soldiers and take them out, without armor and without a lot of back-up. As
he moved up the chain, he was offered a chance to go to ba alion
commander courses. In 1997, he told the army no thanks. He would serve
as a tank hunter in the reserves, but he wanted to get back to the
technology that helps give Israel a leg up over its enemies.
His next stop was an intelligence technology unit. It all started with an
interview with the head of the electro-op cs division. “I remember he
asked me, ‘You finished Talpiot four years ago; now what do you want to
do?’ I said I didn’t know. I knew I wanted to be back in research and
development in technology. He took out a very small camera and said,
‘Anything that you find interes ng here?’ ‘Oh, I like cameras,’ I answered.
‘In my an -tank unit I was actually working with cameras: signal
processing, electro-op cs and cameras.’”
It was a perfect match. He had the formal educa on and the army field
experience to help design what was needed for other combat troops. This
was actually exactly how Talpiot was supposed to work. A promising and
mo vated soldier gets an early and excellent educa on. He then hits the
field. A erwards, he combines both to help give Israel access to more
efficient and more lethal weapons, making the army be er and stronger.
“I started out designing a small board with a camera and signal and
video processing unit, then on to larger components and larger cameras
and op cs systems,” he con nues. “The device I was working on was to be
used for special tasks and missions. They were for the intelligence
community, not necessarily the army,” he says slyly. While he would not
confirm it, it’s likely the devices he worked on wound up helping Israel’s
various security agencies that monitor and police the hos le Arab
popula ons living in ci es and towns east of Israel’s major popula on
centers.
“We were making very, very ny devices. They would put them where
ny devices were needed most. These kinds of devices helped many
people to do their jobs. Many of these missions nobody will ever hear
about.”
One of Israel’s most immediate and pressing problems comes from Gaza.
While Gazans aren’t a threat to the overall security of the na on,
thousands of rockets have been fired from Gaza into Israeli civilian
communi es by Hamas, Islamic Jihad and other terrorist groups. Terrorists
have a acked and a empted to break into Israel dozens of mes, and on
one occasion (in 2006) killed two members of a tank unit and took another
soldier, Gilad Schalit, hostage.
In order to discourage border-crossing terrorism, the IDF set up a ring of
surveillance sta ons to protect communi es near Gaza, without having to
engage anyone who appears to be approaching the border in a threatening
manner. Brigadier General Eli Polak, head of the field intelligence corps,
told Avia on Week, “Our job is to provide surveillance along Israel’s
borders. To do this, we use various intelligence, surveillance and
reconnaissance systems which help us track the enemy and assist ground
forces in quickly loca ng a emp ng hos le infiltrators.” Here again, Talpiot
graduates have played an outsized role in helping to develop and install
sophis cated monitoring mechanisms.
Ofir Zohar (fourteenth Talpiot class) served in a technology unit in the
IDF. He says, “The most advanced stuff in our circle was dedicated to
building be er technology for IDF intelligence. It was our job to come up
with solu ons for problems the army thought were impossible to solve.”
Created by a team working on new components for tank units, the
groundbreaking technology known as the Trophy System addresses one
such “unsolvable” problem. Trophy is designed to protect tanks against
rocket-propelled grenades and other deadly and more accurate an -tank
weapons. Israeli defense contractor Rafael, in connec on with the Elta
Group Division of Israel Aerospace Industries, has ou i ed Israeli Merkava
tanks and some armored personal carriers with it.
Trophy has its roots with Professor Azriel Lorber, who taught hundreds
of Talpiot students the art of military technology during his nineteen years
of affilia on with the program. Professor Lorber served in the IDF armored
corps in the 1950s, rising to the rank of major. He received a master’s
degree in mechanical engineering from the University of Pi sburgh and
then a doctorate in aerospace engineering from Virginia Tech. A er his
studies, Lorber moved back to Israel and eventually went on to work for
two major Israeli defense contractors, Israel Aircra Industries (which later
changed its name to Israel Aerospace Industries) and weapons-maker Israel
Military Industries.
Though originally rejected, the idea for the Trophy system was later
adapted, modified and finally brought to frui on by Rafael. While the IDF
was ini ally reluctant to install the Trophy system because of its cost, the
Second Lebanon War in 2006 made it clear it had to move forward. Fi y-
two Israeli Merkava tanks were hit by an -tank missiles fired by Hezbollah.
Israeli military leaders came to believe that the next war would be against
a tougher, stronger, larger army that would put its tanks in even greater
danger. If this is what Hezbollah could do, they didn’t want to see what
would happen if the IDF suddenly had to fight Hezbollah, the Lebanese
Armed Forces, Syria, Hamas and perhaps even fighters from other fronts,
all at the same me.
As Lorber had originally planned back in the 1980s, the tank has an on-
board warning and radar system that is ac vated by incoming projec les.
Those projec les are iden fied, and then a shotgun-like firing mechanism
shoots a defensive projec le in buckshot form. The goal is for that
defensive projec le to spread out its fire, connect with the incoming
projec le and then force it to prematurely explode before hi ng the outer
shell of the tank.
In June 2012 the Jerusalem Post reported that State Comptroller Micha
Lindenstrauss had heavily cri cized the minister of defense and the IDF for
not expanding the use of Trophy faster to protect more tanks, armored
vehicles and especially the Namer armored personnel carrier.
During Opera on Protec ve Edge in July and August 2014, Trophy got its
first ba le test. It successfully detonated and destroyed a Hamas an -tank
rocket, saving both the tank and the crew inside. The army has been ght-
lipped about the details of that first successful combat use of Trophy, but in
no uncertain terms a spokeswoman from the IDF says, “It has now been
proven to be successful in combat.”
The Israel Military Industries company, also known simply as IMI, has
developed “Iron Fist” closely based on technology used in the Trophy. It is
stronger than the Trophy in that it is capable of deflec ng more powerful
tank shells, not just the hand held an -tank weapons the Trophy is capable
of defea ng. While the Israeli Ministry of Defense did approve usage of
“Iron Fist” in 2009, that decision was later overturned; and as of now, the
technology and the know-how behind it has been put on ice.
While the Iron Fist and Trophy systems are designed to protect Israeli
soldiers in ground combat, usually not far from Israel’s popula on centers,
the long arm of Israel is the Israeli Air Force. It can strike without warning
all over the Middle East and well into Africa. In recent years, western
media reports say Israeli pilots have been called upon to hit targets
carrying Iranian weapons moving throughout Africa, Syria and Lebanon, as
well as weapons-making plants in places as far away as Khartoum, Sudan –
eleven hundred miles away from air bases in southern Israel.
A er his Talpiot academic career at Hebrew University came to a
successful conclusion, Marius Nacht went on to work in aerospace. He
helped design and manufacture airborne systems for the Lavi fighter jet.
At the me, the Israeli-made Lavi rivaled the F16 and the MIG. But there
were problems. First off, it was very expensive. Should a country with only
six million people spend hundreds of millions of dollars on making a fighter
jet? Or would it be more cost effec ve to take the money Israel gets from
the United States (a er signing peace agreements with Egypt and Jordan,
who also get US money for defense for signing those trea es) and buy
proven, fight-and flight-tested American planes?
A second big issue was pressure from the American government against
the project. The United States didn’t want to compete with the Lavi in the
lucra ve interna onal defense market if at all possible.
Israel always has been nervous about its reliance on other countries for
defense. A er the Six-Day War, France – Israel’s main supplier for fighter
aircra – suddenly decided it was be er to align itself with the Arabs than
with Israel. France had been supplying Israel with Mirage jets made by
Dassault. When Charles De Gaulle and France turned their backs on Israel,
it was le with a true security crisis. Where would it find airplanes?
Fortunately for Israel, the United States quickly stepped in to fill the void,
as President Lyndon Johnson saw in Israel an ally that could be a check
against Soviet aggression in the Middle East.
In part because of the trauma caused by the French, and because of
Israel’s exper se in aerospace, it decided to move forward with the Lavi
project. Several Lavi aircra were produced by Israel Aerospace Industries.
The maiden test flight took off on December 31, 1986. Reports say the
plane was remarkably responsive and maneuverable in the air, fast and
smooth. But in the end, Israel’s government believed that building its own
fighter was neither economical nor poli cally expedient, so the project,
while successful, was halted.
Nacht says, “When word came in the Lavi had been cancelled, I was
upset. It was a phenomenal fighter and it could have been a game changer
for Israel. Yet, at least many of the systems that are being used now are
based on the system we developed on the Lavi back then. On a je ighter,
everything must be interconnected. There were many advanced concepts
regarding interface. Now they’re the standard, but back then they were on
the edge. If we’d had to go to war, they would have made a huge
difference.”
A good por on of Nacht’s work on the Lavi was on-board in-flight missile
defense. “It was a very innova ve and crea ve way of protec ng airplanes
from missiles. As far as I can tell, it’s s ll not being deployed. The
Department of Defense in the US now knows everything about it, but I
think that system is s ll ahead of its me. There might be reasons why it is
not being deployed now; there must be a good reason, but I don’t know
it.”
Many of the things Nacht worked on, including airborne missile defense
systems, were later adapted for use on Israel’s fleet of F15s and F16s. Israel
has a special contract with the American manufacturers of the
fighter/bombers. In essence, Israel is allowed to install some specially
designed Israeli components for communica ons, missile defense and
radar. Intelligence es mates say that Israel has about seventy-five F15s
made by Boeing and about 330 F16s manufactured by General Dynamics,
all of which have Israeli designed and Israeli manufactured electronic
warfare systems that advanced rapidly during and a er the work done on
the Lavi by engineers like Marius Nacht.
Similar arrangements have been agreed upon between Israel, the United
States and Lockheed Mar n, which builds the F35. All of the new F35 jets
arriving in 2015 and later will have advanced Israeli electronic warfare
systems. In addi on, Lockheed Mar n also agreed to buy about four billion
dollars of equipment from Israeli defense contractors to install into the
body of the advanced fighter/bomber.
Another Talpiot grad, Amir Peleg, worked on targe ng mechanisms for
Israeli F15s and F16s, though his primary work involved the research and
development of high-tech cameras that could go on UAVs and tell the
difference between different kinds of targets. “More specifically,” says
Peleg, “we built computer-driven vision devices that allowed for automa c
target recogni on. You want a gun to be able to dis nguish between a tank
and a car. We worked on things that are s ll in use in this field.”
Zvika Diament is a rarity for Talpiot. He wears a kippah and is religiously
observant. He is one of the few students to have come to the program
from a yeshivah rather than from a secular high school.
During the interview por on of Zvika’s tryout for Talpiot’s sixth class in
1984, he was asked “How does an airplane work?” With a grin he says, “I
knew that one.” That ques on was prescient. A er finishing his Talpiot
coursework with the equivalent of three majors – physics, computer
science and mathema cs – Zvika went to work on installing and integra ng
Israeli-made electronic warfare components that were to be added to
Israel’s new and growing fleet of F15s and F16s.
He was the Israeli Air Force’s representa ve inside a defense company
called Elisra (now a unit of Israeli defense contrac ng giant Elbit). During
the five years he worked there fulfilling his commitment to the army, Zvika
was involved in every aspect of development in new systems. He was just
twenty-one years old when he started there, and Elisra was filled with
more senior engineers who weren’t always on the same page with what
Zvika or the air force wanted. He notes, “It was very difficult, unpleasant at
mes. They were from a me before Talpiot, and they did things
differently. They largely knew about Talpiot from news ar cles, but had no
actual experience with Talpiot graduates in the workplace. Some were
nice. Some were nasty and tried to get rid of me.
“I was sta oned in the offices of the contractor. I had to define the
acceptance tests for systems, and for each stage to make sure they were on
track. I was in all the mee ngs, trying to provide them with solu ons when
we got into disagreements. And there were a lot of disagreements: They
wanted to deliver what they had so they could get the money from the
army, but some mes what they wanted to deliver wasn’t what we wanted.
Over the years, they learned that I was sent by the air force and they had
no choice but to accept me. The air force backed me up every step of the
way, all the me, so they learned to deal with it.”
Once the electronic warfare parts were made and ready to go, Zvika
would lead the tes ng process. He o en worked with air force pilots who
had also studied to be engineers. This way they could act as both a pilot
and engineer, determining what worked and didn’t work, and why, during
test flights.
Some pilots in the Israeli Air Force do their service, for about five years,
then move on, but later serve in the reserves. Zvika’s testers were some of
Israel’s most experienced pilots; many had fi een or more years of flying
experience. That was especially useful because they were very helpful
when it came to looking at the issue from macro and micro points of view.
Zvika explains, “Let’s say there was an experiment of a missile coming from
one side, but we wanted the bigger picture. You take the aircra and turn it
180 degrees, then 360 degrees, so you can see the level of the signal on all
sides – where the signal comes in high, where it comes in low and where it
is not iden fiable. The pilot has to have a deeper knowledge in order to do
the test perfectly. He needs to go beyond what he used to do to get rid of
enemy aircra or incoming surface-to-air missiles in combat. He needs a
deeper understanding of what was working and what wasn’t. That
knowledge will save lives later on in a real fight.
“What we were doing is working with signal processing. You get the
radar signal in your receiver, then you analyze the signal to iden fy what
kind of missile is threatening you, a SA-6, a Patriot, whatever. For each kind
of different missile, you react differently. For some, you transmit loud
electronic noise to cause the missile to miss you. For others, you throw
some flares to deceive heat-seeking missiles. You have to iden fy the
missile threat within a few seconds to give you me to react to the threat.
If it’s transmi ed from another aircra , the pilot has to react within
seconds; some mes, a er twenty seconds the fight is over. During tests we
simulate the signal; we’re not actually firing missiles. Some mes you can
simulate the ba le situa on with another airplane.”
In the late eigh es and early nine es, Zvika was working inside Elisra as
orders were pouring in and Israel took more and more American-made
F15s and F16s. Zvika was also tasked with traveling to the United States to
make sure the specially designed electronic warfare system parts Elisra had
manufactured were compa ble with the F15s and F16s.
This was no mean feat. “General Dynamics (maker of the F16) would not
give us warran es unless we tested our Elisra systems to make sure they
were compa ble,” Zvika recalls. “The systems were two feet by two feet
and you need several for each plane; they are in different places of the
aircra . They tested it as a black box to make sure there were not extra
electronic current demands that could mess with the big picture, or that
you’re not sending harmful electromagne cs onto other systems, or that
you’re not delivering anything that could cause electric shock. They didn’t
mind if our system didn’t alert on the incoming missiles. They only cared
that our system didn’t mess up the plane in a way that could impact the
warranty.”
Intelligence and aerospace are two core components of Israel’s defense
doctrine. If one falters, lives will be lost, as there is always an enemy
wai ng to pounce. Talpiot grads con nue to play a major role on both
fronts, due in large to part to their training. Their mul disciplinary
approaches to complex problems and the ability to master projects that
require teamwork and coordina on are skills crucial to designing fighter
jets and developing intelligence systems.
Former chief of staff Raful Eitan inspects the troops
(courtesy of the IDF Archives)

An early Talpiot computer science class


(courtesy of the IDF Archives)
Professor Novik teaches a Talpiot class advanced mathema cs
(courtesy of the IDF Archives)

Felix Dothan, founder of Talpiot


(courtesy of the IDF Archives)
Gathering of an early Talpiot class
(courtesy of the IDF Archives)

Beached ships in Angola during the civil war


(courtesy of Gilad Lederer)
Gilad Lederer and crew on his missile boat
(courtesy of the IDF Archives)
Gilad Lederer as a bridge commander
(courtesy of Gilad Lederer)

The third class of Talpiot celebrates


(courtesy of the IDF Archives)
Talpiot cadets train on a Merkava tank
(courtesy of Avi Poleg)

Talpiot cadets receive explosives training in Lebanon during the First Lebanon War
(courtesy of Avi Poleg)
Talpiots Gilad Lederer and Avi Fogelman rest in Lebanon during the First Lebanon
War
(courtesy of Avi Poleg)
David Kutasov, Talpiot graduate, in Lebanon a er the opening stage of the First
Lebanon War
(courtesy of David Kutasov)
Talpiot cadet Avi Poleg (who would later command Talpiot), hand raised
(courtesy of the IDF Archives)

An early Talpiot class prac ces a beach landing


(courtesy of the IDF Archives)
Amir Peleg at the World Economic Forum in Davos, Switzerland
(courtesy of Amir Peleg)

Talpiot’s fi h class in 1985


(courtesy of Amir Peleg)
Elad Ferber outside the gates of Auschwitz on a Talpiot class trip
(courtesy of Elad Ferber)

Gilad Almogy assembling a solar dish at his company, Cogenra


(courtesy of Gilad Almogy)
Gilad Almogy discussing solar power with former United Kingdom
prime minister Tony Blair
(courtesy of Gilad Almogy)
Gilad Almogy with venture capitalist Vinod Khosla
(courtesy of Gilad Almogy)

Detail and depth of a tunnel dug by Hamas stretching from the Gaza Strip into Israel
(courtesy of the IDF Spokesperson’s Office)
An Iron Dome ba ery in ac on during Opera on Pillar of Defense in November 2012
(courtesy of the IDF Spokesperson’s Office)

Ra’anan Gefen (right), working in the lab


(courtesy of Ra’anan Gefen)
Ra’anan Gefen in the Negev desert for an ar llery exercise with Talpiot classmates in
1981; le to right: Shlomo Dobnov, Opher Kinrot, Yuval Yehudar, Ra’anan Gefen
(courtesy of Ra’anan Gefen)

Former chief of staff Benny Gantz and Defense Minister Moshe Ya’alon a end a
ceremony for soldiers of Unit 8200 in July 2013
(courtesy of the IDF Spokesperson’s Office)
Adam Kariv (second from le ) and cadets from Talpiot’s eighteenth class pose in the
Negev during a training exercise
(courtesy of Adam Kariv)
Ophir Kra-Oz working on a chemistry experiment as a Talpiot cadet in 1992
(courtesy of Ophir Kra-Oz)

Ophir Kra-Oz and team members from EMC in Israel outside their Beer Sheva office in
2011
(courtesy of Ophir Kra-Oz)
Members of Talpiot’s ninth class on tank maneuvers in 1990; le to right: Oded
Govrin, Guy Bar-Nachum and Guy Levy-Yurista
(courtesy of Guy Bar-Nachum)
Talpiot cadet Matan Arazi in sniper training
(courtesy of Matan Arazi)
Israel launches the Ofek 7 on June 11, 2007
(courtesy of the IDF Archives)
CHAPTER 12
TALPIOTS IN SPACE

A er Marina Gandlin graduated from the academic por on of Talpiot,


she gravitated to research and development in Israel’s aerospace industry.
When missile a acks from Gaza once again began to escalate in December
2008, she was tasked to improve Israel’s early warning alarm system so
that people in Israeli communi es close to the border with Gaza could
have a decent chance of finding cover before impact. It was cri cal to
design ways to make the alert system work faster: If there is only a thirty-
second warning and it takes a full four or five seconds to detect a launch
and determine the missile’s direc on, saving two or three seconds could
save someone’s life.
Marina explains, “My branch had responsibility for watching the skies of
Israel – planes, air traffic control and everything else that goes through the
sky. Since we have the radar for planes, we also have radar for missiles. I
handled a lot of the launches and the hit points, which is very important
because it lets you know which civilian popula on you need to alarm
before the rocket hits. We were trying to help the army figure out where
projec les are going. I needed to determine where the missiles were going
to hit and then quickly relay the coordinates. It was my responsibility to
decide how many areas and how many sectors to alert a er a launch.”
A er a series of launches, Marina would take the data from missile
launches and where they landed, and enter it into a special computer
program designed to help the army learn from past a acks. This could help
determine likely strike areas in the future.
A er a new cease-fire went into effect January 18, 2009, Marina began
planning her next step, working in Israel’s satellite industry. In 1988, Israel
launched its first satellite, the Ofek (Hebrew for horizon). That successful
launch turned Israel into the eighth na on to have a na ve satellite launch
capability. Since then, nine Ofeks have been launched into outer space
from Israel. It’s believed the Ofek satellites make about six passes over
earth each day. Israel also uses the Amos series of satellites, launched from
the territory of other na ons, o en former republics of the Soviet Union.
The Amos satellites are usually for communica ons purposes. The Ofek is
not.
Its purpose is to take high resolu on image photography of any place on
Earth. The Israeli army and other intelligence agencies are usually the
beneficiaries of what the Ofek satellites send back from outer space.
Gandlin’s job in the Ofek program is to squeeze out every piece of data
the satellites pick up. She develops algorithms to take advantage of the
treasure trove of data and images that comes home. Marina laughs, “Don’t
think I’m looking at license plates or faces. In the satellite industry, people
are laughing at this concept of being able to see faces and license plates.
There’s this movie with Will Smith, ‘Enemy of the State.’ You have real me
images; they follow him with the satellites, you can see if he’s si ng,
standing or smiling. You really can’t get that from satellites.”
While Israel’s satellite program does not live up to Hollywood fantasies,
Marina says she’s coming up with programs that help the army and
intelligence agencies find whatever they’re looking for, whenever they’re
looking for it.
While Gandlin’s ambi on was to be part of the aerospace industry, Kobi
Kaminitz had no clue where his true interest was, un l a rou ne class trip
turned his life around. Quiet and though ul, Kobi’s Talpiot commanders
had high hopes for him the minute he was inducted into Talpiot’s sixteenth
class. They believed he could make a difference in Israel’s future on a grand
scale.
Kaminitz had other plans. He wanted to go into the army to become a
fighter, a commander of troops in the field. He was all set to tell his Talpiot
superiors – who had invested so much in his educa on – that he wanted to
enroll in a field commander’s course. It’s not clear how they would have
reacted and it doesn’t ma er. He never had that conversa on.
One day, his Talpiot class was on a field trip to see parts of the Israeli
space program in the works. When they went into a hangar and he saw the
satellite Ofek-4, “that was really magic for me,” he recalls. “I vividly
remembered watching the launch of Ofek-3 on television. Then I was able
to see another Israeli rocket launch in person. I knew then this is what I
wanted to do. It was fascina ng. You see it on the ground one minute, then
a few minutes later it’s four hundred kilometers in space. I knew I wanted
to be part of that.”
A er his third year of Talpiot coursework, he spent the next six years
working on the camera for Ofek-5. At the age of only twenty-one, he was
working on a $100 million project crucial to his country’s defense. When
asked how he was able to get this top priority post, Kobi was extremely
modest, saying, “Someone somewhere thought I was a good mul tasker
and was easy to work with.” Of course, both of these quali es are
extremely valuable, but Kobi had a lot more than that: extreme mo va on
to help his country, a desire to work hard and a world-class educa on in
military technology.
On May 28, 2002, the Associated Press in Jerusalem reported: “The
Ofek-5 reconnaissance satellite was successfully launched and will soon
begin providing Israel with high-resolu on images of the Middle East.
Defense Minister Binyamin Ben-Eliezer said that the satellite and its
launcher rocket, developed en rely in Israel, were ‘a tremendous
achievement for the Israeli defense establishment.’ The Ofek-5 satellite,
developed by Israel Aircra Industries (IAI), rocketed into space onboard a
Shavit launcher at 6:25 pm, from the Palmachim air force base. The Shavit’s
engines shot out a huge white vapor trail as the rocket sped westward over
the Mediterranean. Minutes later the rocket and satellite disappeared over
the horizon.”
Looking back, Kobi says, “I remember the clock cking down, from nine
hours down to minutes, down to seconds. I was tes ng the camera to
make sure that everything was okay in the monitor I was looking at. There
are also a lot of different configura ons you can work with. Exposure me,
the shu er; it is just like your camera. The camera on the Ofek is a very
complex equivalent. I could tell it to take pictures of this point on Earth,
that point on Earth. You switch and test to make sure you can see the
inputs – of which there are several dozen – giving you different views. To
play a part in this was really a great experience. When the first image from
the satellite came in, there was nothing I could compare that to.”
He then worked with demanding Israeli intelligence agencies a er the
launch to provide them with whatever they wanted. A er a short me,
Kaminitz was able to provide them with images without being asked, as he
was able to figure out what they needed. Intelligence officers would o en
bring him pictures of other targets, asking him to check for updates on
everything from weaponry to troop movement to the placement of tanks
and missile launchers throughout the Middle East. The intelligence teams
also had instruc ons when it came to op mizing the pictures of what they
wanted to see, and they were very specific in reques ng size and shading
so they could be 100 percent sure of what was happening on the ground.
A er working on the Ofek-5 camera, Kobi wound up leaving the army
saying it would be difficult for him to find something as inspiring to work
on. (Not long therea er, he wound up using the skills he had acquired to
work on very similar technology in the private sector to help pa ents in
danger of losing their eyesight.)
Israel’s space program and electro-op cs are the exper se of Tal Dekel,
of Talpiot’s seventeenth class (1995). Dekel is currently a research fellow at
Tel Aviv University’s Yuval Ne’eman Workshop for Science, Technology and
Security, which was founded by General Yitzhak Ben-Israel (the outstanding
individual whom we met in chapter 7). The department focuses on a wide
range of security issues through the prism of science. On the list is cyber-
security, Israeli space policy, guided weapons, ballis c missile technology,
nuclear energy and robo cs to name a few. When it comes to space, the
program focuses on using satellites for improving intelligence-gathering.
While mo ves and inten ons of a foreign leader can be debated, satellite
imagery gives a clear picture of what’s happening on the ground.
Through his work at Ne’eman, Dekel has been called upon to analyze the
progress being made in space by na ons in Israel’s neighborhood including
Egypt, a country many people – even those in security circles – didn’t know
had a space program.
Dekel isn’t impressed.
Egypt says its satellite program is for scien fic use, but many experts
believe a country like Egypt – a na on with a ques onable economy and a
powerful military – wouldn’t spend so much money purely for civilian
purposes. Dekel believes that, as in most countries, their program is for
dual use, military and civilian.
In 2007, Dekel helped monitor the launch of EgyptSat-1. Egypt was able
to reach space with the generous support of scien sts and space experts in
the Ukraine. But by 2010, communica ons were lost with EgyptSat-1, and
dozens of Egyp ans who had worked on the program were fired. Dekel
says the Egyp an government hid the bad news for months. Among his
other responsibili es, Dekel o en represents Israel at interna onal and
United Na ons-sponsored space conferences. In the spring of 2011, he was
in Geneva. He gave a presenta on proposing interna onal rules for the
management and governance of space, rules that every na on would need
to abide by to prevent one na on from interfering with another beyond
Earth’s atmosphere.
One example where space rules are needed relates to jamming signals.
Many countries have the ability to stop signals from coming into their
countries. (Dekel notes that Iran is actually a world leader in jamming
technology and there’s no way to prevent it. Dekel says you can retaliate by
jamming their signals, but in the end, nobody in that scenario really wins:
all that money that went into satellite launches and the sending back of
signals is lost.) Beyond jamming, it is also possible to blind satellites with
certain kinds of lasers – another area where interna onal rules are badly
needed.
A remarkable thing happened when Dekel presented his proposal in
Geneva. Almost always, Iran’s representa ves – and the representa ves of
many Arab na ons – boyco speeches given by Israeli experts at
interna onal conferences, including conferences run by the United
Na ons. But to Dekel’s recollec on, when he spoke, it was the first me
that the representa ves of Iran did not leave the room.
CHAPTER 13
MISSILE COMMAND

W e first met Ophir Shoham in chapter 8 when the Talpiot cadet


launched a six-foot, two-hundred-pound paratrooper ten feet in the air for
picking on one of his classmates. Since then, Shoham became the highest
ranking and most prolific graduate of Talpiot in the military.
He rose through the ranks of the Israel Defense Forces, the navy (where
he commanded a missile boat) and the Ministry of Defense to become a
brigadier general in the reserves and the head of MAFAT, the
Administra on for the Development of Weapons and Technological
Infrastructure. General Shoham also holds a seat on the all-powerful
General Staff.
In this posi on, he became responsible for pushing forward missile
defense. Israel now has defenses capable of shoo ng down three types of
missiles:
The Iron Dome hits short-range rockets. It was first deployed in 2011,
but it became famous during Opera on Pillar of Defense in 2012.
David’s Sling, which is not yet fully opera onal, but is ready to go. It is
designed to take down missiles fired from 18 – 180 miles away.
David’s Sling is also some mes known within Israel as Magic Wand.
The Arrow System, the most complicated and perhaps the most
important, was developed to defeat long-range ballis c missiles that
could be fired from launch sites in Iran.
Shoham and Talpiot graduates have been essen al in the technology
behind each and every one of them. It was a perfect project for a Talpiot
graduate, and many Talpiots inclined to this area of Israel’s defense.
Missile defense is a complicated project that requires a major mul -
discipline approach. Designers need to take into account the highest level
mathema cs, physics, radar detec on, propulsion, explosive charges, the
vehicle itself, along with communica ons – to name a few of the required
disciplines necessary for successful missile defense. Then appropriate areas
for deployment, where the an -missile ba ery will have maximum impact,
must be determined. And the people working on all those different areas
must get along well enough to move the project forward together, quickly.
According to a professor who was involved with Talpiot’s student projects,
the idea for Iron Dome was first dreamed up by a group of cadets in the
1990s, when Hamas began firing short-range homemade rockets and
mortars at Israeli communi es in the Gaza Strip. Projec le a acks on
communi es such as Gush Ka f and Dugit became more and more
frequent.
The rockets were very primi ve, more like mortars, and while they
weren’t doing a lot of physical damage, the psychological effect on the
popula on under a ack was becoming significant. Re red Colonel Shaul
Shay, former deputy head of the Israel Na onal Security Council, says, “The
government didn’t consider these a acks a serious threat, but if you lived
in Gaza and your home was coming under a ack, you wanted it stopped.”
The ques on was how.
Hamas’s rockets, later called Qassams (named for Izz ad-Din al-Qassam,
who made a name for himself a acking Jewish people living in the Haifa
area in the 1930s) became more sophis cated and more dangerous over
me. From the mortar-like ordinance in the 1990s, they evolved into larger
and more aerodynamic missiles, some mes made with the long metal
canisters found inside traffic lights, stuffed with explosives, nails and ball
bearings. A er the rockets were fired, the launchers could be easily and
quickly hidden.
The IDF wasn’t sure how to stop the a acks. They couldn’t cover enough
of the Gaza Strip from the air or by land to catch the men firing the rockets
red-handed. A er a missile firing, the terrorists most o en blended back
into the civilian popula on and hid their launchers, some mes in homes,
schools or mosques, and more recently in sophis cated underground
bunkers.
It was a cat and mouse game between the IDF soldiers sta oned in Gaza
and small groups of nimble Hamas terrorists who knew the topography
and ci es be er than Israeli troops knew them. Israel didn’t want to send
soldiers in to a ack these small nests of terrorists every me there was a
rocket launch. First off, it wasn’t an efficient method. Secondly, most o en
they’d be playing on Hamas’s terms and on their territory. Most
importantly, the IDF would suffer casual es. In Israel, it’s a front-page story
when a soldier is wounded; it’s the lead story on television news, and the
lead story on news radio, which airs on almost every sta on in the na on
at the top of each hour. Everyone in the country hears the details quickly
and cares about him.
Hamas also had other ways of a acking Israelis. They le roadside
bombs; they periodically inflicted shoo ng a acks, and star ng in the early
1990s, they were able to terrify Israeli ci zens with suicide bombings.
These a acks really became a serious problem for Israel in 1994, when
they almost exclusively targeted civilian buses, nightclubs, cafes, bars and
restaurants, killing dozens of Israelis at a me and injuring even more. For
the first me in quite a while, Israelis were running scared and were
constantly looking over their shoulders.
A er watching the escala ng rocket a acks for a few months, a group of
Talpiot cadets decided to focus their second-year class project on a fairly
low-cost solu on to stop Hamas rockets. It was a clever presenta on, and
it caught the a en on of some of Israel’s top military research and
development officers. But the idea didn’t go far in its first itera on. The
military was not ready to implement such as system. But the Talpiot idea
and the prototype they developed was a start.
In 2002, Ophir Shoham was appointed head of planning in the IDF. Along
with the head of MAFAT research and development, General Daniel Gold
(the man largely credited with developing the modern itera on of Iron
Dome), Shoham pushed the missile defense system as a solu on to the
short-range missile threats. Engineers at the Ministry of Defense then
began taking the concept of Iron Dome more seriously, believing it could
be a realis c solu on to a threat they correctly believed would only
become worse in the years ahead.
Their plans ran into major obstacles from the army. Many generals
argued vociferously that it was the army’s job to bring the war to the
enemy, to be on the offense, not to spend money on unproven defensive
measures. Figh ng belongs on enemy territory, they persisted, not in
Israel.
But in 2005 Prime Minister Ariel Sharon shocked much of Israel. The
man known for “breaking the bones” of Arabs who threatened Israel
uprooted the eight thousand Jews living in Gaza and forced them to move
inside Israel’s interna onally recognized borders, or to Judea and Samaria,
areas known in the West as the West Bank. Many in Israel and the world
hoped this would be the first step toward peace with the Pales nians.
Sharon himself called it a test to see if the Pales nians were really ready
for peace; he wanted to see what would happen if they were given
autonomy. The world quickly found out.
Hamas violently took over the Gaza Strip, destroyed infrastructure and
greenhouses le behind and turned Gaza into a launching pad for Qassam
rockets – and later for longer range, more sophis cated weaponry like
Russian made and Iranian supplied Katuysha rockets that could reach
further inside of Israel.
Between 2001 – when the Qassams were first fired – and 2012, there
have been more than ten thousand rocket a acks against Israel fired from
Gaza. More than 90 percent of those a acks came a er Israel le Gaza.
Despite objec ons from many army generals, Defense Minister Amir
Peretz (one of the least popular defense ministers in Israel’s history)
became an early backer of funding Iron Dome when he took office in 2006.
For Peretz, it was an easy decision. He grew up in Sderot, which had been a
constant target of Hamas rockets, and was extremely sympathe c to
people forced to endure the dangers and disrup ons to daily life caused by
missile fire from Gaza. It was his final decision that allowed more funding
to start flowing into the project.
In an in-depth ar cle in The Times of Israel, Mitch Ginsburg reported, “In
February 2007, with funding for just one year and without the requisite
signature of the finance minister necessary for all mul -year projects,
Peretz authorized the development of Iron Dome. At a midnight mee ng in
his office…he reached an agreement with Rafael defense systems officials:
They’d ‘scrape together’ $50 million and the Defense Ministry would
‘scrape together’ another $50 million – out of an annual budget of some
$15 billion – and produc on would start immediately.” The engineers
working on the project were able to show the Ministry of Defense a
working model three years later.
In the mean me though, the rockets con nued to fall. The a acks
forced Israel to launch Opera on Cast Lead at the end of 2008 through the
beginning of 2009. Throughout the figh ng, Hamas and other groups
launched hundreds of rockets at civilian areas. The government was under
pressure to find a solu on to protect civilians from the constant threat.
Israelis in the communi es within rocket range of Gaza complained that
if the rockets were falling on Tel Aviv or Jerusalem, the government would
have an answer: the army would retaliate harshly. There were calls for the
government to reinforce the roofs of homes, schools and community
centers to protect civilians. Many Israelis knew of the work being done on
the Iron Dome due to news reports, and there was intense public pressure
to set it up, fast.
In March 2011, the Iron Dome system was first deployed in the Negev
desert to protect ci es and towns closest to Gaza. The Ministry of Defense
made the move with li le fanfare, but a lot of internal debate. The system
had not been fully tested: it was called a “hot rollout” to further downplay
expecta ons.
On the early rollout, Brigadier General Shoham was quoted in the Israeli
daily Haaretz saying, “Senior defense officials were right when they made
this decision; at the Administra on for the Development of Weapons and
Technological Infrastructure, we felt it would be proper to go ahead, even if
the air force had reserva ons. It is not that they opposed this, but they
were delibera ng. We could have not deployed the system, absorbed
casual es, and the poli cians would have had less room to maneuver. In
such a situa on, we could have absolved ourselves by claiming the system
simply was not ready. But the direc on was clear and there was an
excellent combina on of opera onal planning and technological and
logis cal support. In my understanding, we were ready enough to take a
calculated risk here, and it is not all that great a risk, because the trials
were 100 percent successful.”
Iron Dome is made of three main parts: a radar sta on, the control
center and the interceptor ba ery which launches the an -missile
projec le. A number of Israeli defense contractors are credited with
manufacturing the system, including the biggest names in the Israeli
defense industry – Israel Aerospace Industry, Elta and Rafael, as well as
many of their subsidiaries.
The radar, of course, tracks incoming missiles. But it goes a step further,
telling the command crew which incoming missiles pose a threat to
communi es and which ones don’t. Firing an an -missile from Iron Dome
is a rela vely expensive endeavor. Hamas is able to make Qassams for a
few hundred dollars a pop. The an -missile projec les cost Israel about
$38,000 every me one is launched, so having the system prevent waste is
important. But Israeli officials are very quick to point out that despite the
discrepancy in cost between what Hamas fires and what Israel pays to
combat those missiles, if the incoming missiles hit cars, homes and
neighborhoods, the cost of repairs would be much higher than $38,000, so
you can’t compare the cost, missile to missile. They also point out that
$38,000 is obviously more than a fair trade when it comes to saving
somebody’s life.
Shoham also told Haaretz, “We have no pretension of intercep ng
thousands of missiles, only of gaining me, limi ng the threat and in the
mean me, the army is also doing other things. We must not forget that the
system also contributes considerably to Israel’s deterrent capability.”
The first real test came on April 7, 2011, when a Russian-made Grad
Rocket capable of traveling twenty-five miles was fired from Gaza in the
direc on of Beer Sheva, the largest city in southern Israel, home to
200,000 people.
Those who witnessed history being made say the Iron Dome’s radar
picked up the incoming missile. The system tracked it, and within
milliseconds determined it would likely hit a populated area. Sirens wailed,
lights flashed, the young men and women in the command center began
shou ng commands and following orders. The an -missile was launched.
With a short, loud explosion, it shot into the air. Three seconds, four
seconds, five seconds…ten seconds later an explosion could be seen in the
sky, followed by a loud boom. Success. At a cabinet mee ng a few days
later, Prime Minister Netanyahu said, “Israel marked a significant and
impressive achievement with the Iron Dome system intercep ng missiles.
This has echoed throughout the world.” More successful intercep ons
would follow in the weeks and months ahead.
The army reports that during the April 2011 escala on with Gaza, the
Iron Dome had a 65 percent success rate. That was the month the system
began opera ng. In August 2011, during another round of missile a acks,
the Iron Dome’s success rate hit 70 percent. In March 2012, Iron Dome had
an 80 percent success rate. In June 2012, the Iron Dome’s hit rate reached
85 percent. In October of that year, the success rate hit 95 percent.
More rocket a acks, and Hamas’s acquisi on of missiles that could hit
Tel Aviv, led to a new round of intense figh ng in November 2012, with
Israel launching Opera on Pillar of Defense. During Opera on Pillar of
Defense, 1,506 rockets were fired at Israel from Gaza. The Iron Dome
system determined that 421 of those rockets were a threat, and 85 percent
of those threats were destroyed in mid-air. The Ministry of Defense says it
does not keep exact sta s cs on successful launches, but they do believe
that with some tweaking, the system can be close to 100 percent effec ve
in the years to come. There is op mism within Israel’s defense community
that the Iron Dome will be sought a er by other countries and will add to
the Israeli-made weapons systems the country sells and exports. As of
winter 2015, no agreements have been signed.
While Talpiot originally dreamed up the system and the highest ranking
Talpiot grad, Ophir Shoham, was par ally responsible for making it happen
(with the help of the Israeli military industry), the United States was a big
factor in paying for Iron Dome. Congress approved about $500 million in
funding for the project; and in June 2013, lawmakers tacked on another
$15 million in funding, in the hope that the US could collaborate directly in
future developments. That same month, the House of Representa ves’
Armed Services Commi ee approved $284 million to help pay for
programs designed to team up with Israel in the field of missile defense.
Israel and the United States have been working together on longer range
missile defense systems for decades. The two jointly developed the Arrow
an -missile system.
The Arrow is now in its third genera on and is mainly designed with the
threat of Iran’s ballis c missile program in mind. It is considered to be the
world’s best long-range missile interceptor.
Produc on of the Arrow began in 1986, the same year Talpiot students
were star ng to make a name for themselves in Israel’s space program.
Israel’s space scien sts made a huge contribu on to the Arrow because, in
effect, it is a rocket. The Arrow is tasked with streaking approximately thirty
miles above the earth’s surface, seeking out incoming missiles and then
exploding, knocking out the enemy projec le. One of its many goals is to
destroy an incoming missile armed with nuclear, biological, or chemical
warheads far enough into the sky that the deadly payload doesn’t hit
Israel.
The Arrow is jointly developed by Boeing and Israel Aerospace Industries
at a cost of about $3 million per missile. The rockets run on solid fuel
rather than more vola le liquid fuel. Solid fuel also gives the Arrow missile
commanders the luxury of not having to deploy the rockets shortly before
they’re used; in other words, they’re always ready to go on the launch-pad.
That’s key because missile defense o en can’t be planned. An interceptor
like the Arrow is only used when an enemy fires its missiles, so ming is
unpredictable, to put it mildly.
Like the Iron Dome system, the rocket launchers don’t operate
independently. Instead, there is a command center, a separate radar
sta on and the ba ery that actually fires the missiles.
The Arrow 1, 2 and 3 have all been successfully test-fired in actual
simula ons, and the system has proven it can take out enemy missiles. In
test phases, physicists, engineers and Israeli soldiers manning the Arrow
ba eries claim the Arrow has a 90 percent success rate.
Many in the world of aerospace, including some Israeli experts, have
long doubted the effec veness of the Arrow. Some in the Israeli defense
establishment have even said the billions of dollars spent on the Arrow
system would be be er used elsewhere. Others say the Arrow might work,
but could be defeated with decoys. That is to say if an enemy country fired
fi y missiles at Israel and only one was armed with a nuclear warhead,
with the other forty-nine being decoys, the Arrow would not be able to
save the day.
One Israeli physicist with in-depth knowledge of the capabili es of the
Arrow, who also teaches in the Talpiot program, laughs off those fears. He
agrees missile defense will never be 100 percent effec ve, but decoys can
most definitely be defeated.
Professor Azriel Lorber, a Talpiot teacher for almost two decades, a ests,
“Former students of mine from Talpiot worked on solving the decoy
dilemma in the Arrow’s early years. Decoys show off different flight
characteris cs, giving them away. It is also true that it’s easier and cheaper
for the enemy to build a real missile than to build a good decoy. All of
these factors are helpful in determining which is the real thing and which
isn’t.”
The first Arrow ba ery was installed at Palmachim Air Base south of Tel
Aviv. Several successful test launches and test intercep ons have been
executed from this base. While the Arrow has not been used in a real life
situa on, the officers who would be the ones firing at an enemy missile
have been able to test a lot of their skills during the civil war in Syria.
The Syrian army launched several Scud missiles in their a empt to quash
the rebel uprising that began in 2011. Israel ini ally became very familiar
with Scuds during the First Gulf War, when Saddam Hussein fired three
dozen of them at Israel in 1991.
In 2011, 2012 and 2013, Syria’s firing of Scuds set off alarm bells at
Palmachim Air Base, forcing an -missile teams to have their fingers on the
triggers, as radar tracked those launches inside of Syria.
When Syria fires its Scuds, Israel tracks those missiles. They’re doing it in
part to test their tracking abili es and to study the way Syria fires the
Scuds, but most importantly, Israel tracks them to make sure they’re not
being fired at Israel. The Ministry of Defense says that for the first few
cri cal seconds it is difficult to tell if a acks that go from north to south are
aimed at rebel held areas in Syria or if Israel is the target.
In 2013, Colonel Zvika Haimovich told Reuters news service that a er
Bashar al-Assad’s forces fire a missile, Israel only has a few seconds to
determine it if is the target or not. “Syria’s ba eries are in a high state of
operability, ready to fire at short no ce. All it would take is a few degrees’
change in the flight path to endanger us.” In that same ar cle, Colonel
Haimovich also told Reuters, “We are looking at all aspects, from the
performance of the weaponry to the way the Syrians are using it. They
have used everything that I am aware exists in their missile and rocket
arsenal. They are improving all the me, and so are we, but we need to
study this and to be prepared.”
The Reuters report con nues: “Long-range radars feed real- me data on
the barrages to Haimovich’s command bunker, where officers brace to
ac vate Arrow II, a US-backed Israeli missile shield that has yet to be tested
in ba le. The more threatening launches set off sirens across Palmachim,
whose warplanes also await orders to scramble. Haimovich would not
detail how Israel determines a missile fired in its direc on will not cross the
border, saying only that the process took ‘more than a few seconds, but
not much more.’ Another Israeli expert, speaking on condi on of
anonymity, said it combined split-second analysis of the strength of the
launch with up-to-date intelligence on Assad’s inten ons.”
Israel also has developed a missile interceptor that’s able to detect, track
and destroy medium range rockets heading into Israel. David’s Sling had its
first successful test on November 25, 2012. It is in its final test stages and
hopefully will be deployed in me for Israel’s next inevitable war.
It will be expected to take out rockets fired from up to 150 miles away.
That would put missiles fired by terrorists, or anyone else, from Egypt’s
Sinai desert within range. Over the past several years, dozens of missiles
have targeted the Israeli port city of Eilat.
The system, however, was designed with Syria and Lebanon in mind.
Syria has missiles capable of striking anywhere in Israel. Lebanon is host to
Hezbollah, which boasts it can hit anywhere in Israel with its impressive
missile arsenal. In order to bring about a cease-fire a er the Second
Lebanon War in the summer of 2006, the interna onal community pledged
to prevent Hezbollah from rearming. Despite that, Hezbollah is believed to
have more than forty thousand missiles aimed at Israel and ready to go.
Hezbollah possesses three main missiles, the Zelzal and the Fa eh 110.
Both are made and supplied by Iran. Zelzal is Persian for earthquake, fa eh
means conqueror. The third missile category is the Russian made Katyusha.
The Zelzal and Fa eh-100 are both capable of carrying fi een-hundred-
pound warheads and have an approximate range of 150 miles. That would
put both missiles in the “killable” category for David’s Sling. Syria also
possesses the Fa eh-110, and reports say the Syrian army controlled by
Bashar al-Assad has used them against rebels during Syria’s civil war.
The Katyusha is not as advanced as the Fa eh-110 or the Zelzal, but it
packs a major punch. It was used heavily by Hezbollah during the Second
Lebanon War. It poses a threat to Israel because the missiles are o en fired
from mobile launchers. They can be fired quickly and in heavy volumes,
then hidden from Israeli aircra .
David’s Sling’s is capable of dealing with all three of those missiles, and
its main responsibility will be to prevent Hezbollah’s missiles from
damaging Israeli communi es. With targe ng and guidance devices
implanted in the nose, David’s Sling interceptor missile is some mes
known as “The Stunner.” Compared to the Arrow missiles, the interceptor
fired by David’s Sling is less expensive.
There is no doubt that Israel’s sophis cated missile defenses will be of
cri cal importance in the years ahead, as constant research and upgrading
in this field will provide a much needed sense of security for all of its
ci zens. Its cu ng-edge development and far reaching impact in large part
reflect the work of Talpiot cadets and Talpiot graduates, not the least of
whom is the visionary Brigadier General Ophir Shoham.
CHAPTER 14
ON A MISSION

G iora Kornblau always wanted to fly. When it came me to serve in the


IDF, he was certain he’d shoot for flight school. Giora wanted the air force.
But Talpiot wanted Giora.
As his three years of study started coming to a close, he opted to
postpone working in research and development, the des na on for most
Talpiot graduates, and join the air force. He wanted to be a fighter pilot.
Colonel Avi Poleg, who headed Talpiot at the me, usually encouraged
graduates to consider combat service a er gradua on. When he and his
staff would find a candidate for that kind of route, he would help the
Talpiot graduate navigate the bureaucra c maze leading to combat posts.
“I find the combina on of Talpiot and combat service to be an excellent
way to get involved in major important domains in the defense area,” says
Poleg. “A er serving a period in the field, there is a poten al for building a
long and stable military career, which may lead to the top.”
This was certainly the case for Kornblau, one of the first Talpiot
graduates to go on to flight school. Born in Argen na in 1972, he was
brought to Israel by his family one year later. When Talpiot recruited him,
he had not heard of the program, but he would soon join as part of the
twel h class, inducted in 1990.
When Kornblau announced his inten on to go to flight school, there was
no precedent for making such a move. All the rules were being wri en as
Giora moved forward. Part of his argument was that for three years Talpiot
had pressed upon his class the importance of combat training and the
importance of “ge ng your hands dirty in the field. They want people with
a combina on of an educa on and real combat experience. I agreed with
them.”
He remembers, “There were some bureaucra c hurdles. There are
always roadblocks, but everyone always comes together at the last minute.
I did need to do some convincing at the Ministry of Defense, which had
invested in my educa on. At the end of Talpiot, there is a three month
gradua on project. In order to start flight school a er gradua ng I needed
to miss that. There was a lot of discussion, but in the end the Ministry of
Defense and the leaders of Talpiot said, ‘If you pass flight school, go ahead
into the air force.’ The air force, however, was very black and white about
the whole issue. They said, ‘We don’t need convincing, we just need you to
pass the test.’”
And like almost everything else Kornblau had done to this point in his
life, he was a success. A er passing his ini al tests, the learning curve
steepened. He began learning to fly combat style missions on A4 Skyhawks,
now used for training in the Israeli Air Force. The planes were previously
used by Israel in combat missions in the late 1960s and early 1970s, as well
as in the early 1980s in Opera on Peace for the Galilee. As “flying ar llery,”
their main mission was to pound ground targets and provide cover for
Israel’s ground forces.
A er mastering the A4s, he was taught to fly F16s. The F16 is known as
Israel’s long arm. Israel’s fleet of about 225 F16s can be refueled in mid-air
and have been known to be capable of striking anywhere in the Middle
East to the northern half of Africa, which has become a travel route for
weapons sold to Hamas and other terrorist networks in Gaza and the Sinai.
In June 2008, one hundred Israeli F16s and F15s flew in forma on
toward Greece. The distance is about nine hundred miles, the same
distance from Israel to Iran. Greece is armed with the Russian-made SA-
300 an -aircra system, the same system Russia was reportedly
considering selling to Iran. The point the Israeli Air Force was making was
pre y blunt: it is ready to carry out any mission anywhere, at any me.
The missions of Israeli Air Force pilots usually remain classified for
decades a er they’ve been carried out. Kornblau could only say he was
called on to carry out missions in all the areas where the Israeli Air Force
was ac ve during his me as a pilot. During his me in service, the IAF
carried out bombing missions against Hamas and Islamic Jihad in Gaza,
a er both organiza ons had launched wave a er wave of terrorist a acks
against Israeli civilians. Israel also began targe ng the leaders of those
terrorist opera ons in an effort “to cut off the head of the snake.”
Another theater of opera on during Kornblau’s me in the air force was
to the north of Israel, in Lebanon. He flew during the final months of the
Israeli army’s occupa on of southern Lebanon (which had begun in 1982).
As the Lebanon-based PLO became less of a nuisance for Israel, Iranian-
backed Hezbollah took over as “resistance fighters,” providing the IAF with
targets for years to come.
Since the early days, the fighter pilot has been known as “a thinking
man’s warrior.” You need to know how to operate the complicated
controls, how to not stall in midair; you need to know physics and
aerodynamics. You need to surmise what your opponent is able to do and
when he will do it. Once in the air, Israeli Air Force pilots are given a lot of
leeway to accomplish their mission. Despite his combat experience,
Kornblau admits to being embarrassed when asked what he thinks about
while flying missions. “I don’t regard myself as a brave pilot with brave
stories to tell. Usually I’m thinking of many things. Some are related to the
mission and how to perform it the best I can, simula ng in my mind the
cri cal parts of the mission and what needs to be done. And some mes I
think about lunch, or anything else that pops into my mind. I think it’s the
same for anyone else about to perform something that’s important to
them.”
A er finishing his me as a fighter pilot in the most advanced plane in
the Israeli Air Force’s arsenal, Kornblau took the knowledge he acquired
and moved back to research and development. He said, “The experience
gave me the technical knowledge to be able to work on the future of the
air force – and to help develop the technologies of the future.”
Arik Czerniak is one of Talpiot’s most popular graduates. As a
compe ve teenager, he and his friends were in a constant state of one-
upmanship to see who could get into the best army unit. Czerniak wanted
to win, but he discounted Talpiot, thinking he’d never get in. So he set his
sights on becoming a fighter pilot.
As dra day approached, he was invited to come to Talpiot’s early
tes ng. When he got there, the officers asked him what he wanted to do.
Czerniak was forthright: “‘I want to be a fighter pilot.”
“No problem,” they laughed, “you can do both.”
“They sent me to the commi ee, a tes ng panel,” recalled Czerniak.
“One week before, I had read a book about Einstein’s theory of rela vity in
order to try to look prepared and smart. Everyone needed to prepare
something scien fic to talk about. Then we talked about physics. They
asked me how a solar-powered boiler works, the ones on the roof. They
asked me what I would want to study if I weren’t in the army, and I said
architecture. Then they asked me an architectural problem and I had to
show them how I’d plan a living room. Then they asked me a progression
problem. I even remember it: 61 55 52 63 94…what’s the next number?
“I stood up at the white board and tried all the various mathema cal
methods I knew to solve progressions, and they said, ‘Just flip the order of
the digits,’ and I said, ‘…oh, right.’ I looked like an idiot, but it was funny. It
was just a trick. Looking back, they just wanted to see how well I could
handle pressure.
“At the end of the tes ng I asked again, ‘Can I s ll be a pilot?’ I wanted
to ask as many people as possible, to make sure the answer would always
be yes, and it was. They were true to their word.”
While he was wai ng to hear back from Talpiot, he accepted an
invita on to try out for pilot school. “In the air force, the training was
seven days, with six hundred people. They put you in uniform. You spend
the day running around, comple ng orders. There’s no English word for
what we had to do, but it translates to ‘advancement by the legs.’ You see
that tree? You have twenty seconds to run there and back: GO! You didn’t
make it. Do it again! There were a lot of group ac vi es and tests like
digging holes, solving puzzles, hanging from monkey bars; everyone hangs
and you see who falls first and last. There’s really no sleeping, they woke us
up a er a two-hour rest.”
Czerniak made it past all the hurdles. But in the end he thought to
himself, “Thanks, guys, this is a good failsafe, but now I’m really hoping to
get into Talpiot.” As he graduated high school, he s ll didn’t have an
answer from Talpiot. One day in the early summer, he was playing on his
computer when the call came in. “Congratula ons, you’ve been accepted
to Talpiot’s fi eenth class.” His first ques on – “Can I s ll be a pilot?”
His first day in Talpiot, the commanders brought the cadets pitas and
schnitzel. “Then wham! It was off to paratrooper training. A er six days, I
could barely move my legs, but I kept going and going. It was a piece of
cake compared to what my friends in combat units did, but it was s ll
hard.”
When classes began, the course load was really heavy. But Czerniak had
an ace up his sleeve. He’d always had the ability to concentrate on material
in the hours right before a test and to do well. A er gradua ng from
Hebrew University and Talpiot’s academic program, Czerniak was expected
to do six months of research and development before doing anything else.
It had already been mapped out. His job would be to work on a new Israeli-
made radar system for F16s, systems the Israeli air force would install into
the American-made fighter jets.
But the air base commander where he would serve said, “Enough is
enough. If you want to be a pilot, you have to come now.” The bureaucra c
barriers were crossed, the documents were signed and Arik Czerniak was
off to start army service again from the very bo om. (The Ministry of
Defense later marked this as a bad decision. In the years a erward, with
few excep ons, Talpiot graduates would have to do some work in research
and development before moving on to figh ng units.)
Czerniak was finally learning to fly. A er gradua ng flight school, he was
given the keys to an F4 Phantom Jet, a fighter bomber once used heavily by
the IAF. But soon a er he had been assigned to an F4 squadron, the air
force decided the plane had seen its best years. Czerniak was disappointed,
but happy with his new assignment. To this day, he’s a flight instructor in
an A4 Skyhawk. Like the F4, the A4 was once a big part of the Israeli Air
Force’s figh ng fleet. The flying skills Czerniak learned in those now
outdated planes have s ll been put to great use. “I go to reserves a day or
two every two or three weeks to train pilots in air-to-air combat. I usually
teach dogfights. If tomorrow two F16s – one from Egypt, one from Israel –
would get in close range there would be a dogfight. The dogfigh ng I teach
is like teaching someone to dribble. You’re not firing bullets, of course. Your
goal is to take a picture of the other guy in your gun sights. You’re behind
him, three hundred meters, he’s squiggling in your gun sight, and it’s all
captured on video. When you go down, you debrief to find out who won,
who lost, and why.”
Many of the graduates of Talpiot humbly say no ma er what their
contribu on is to the security of Israel, nothing ever outweighs the
sacrifice made by a real fighter, a soldier on the frontlines, a fighter pilot
flying over enemy territory, or a sailor engaging in ba le on the sea.
Boaz Rippin of Talpiot’s second class knew one incredibly smart young
recruit in Talpiot who came from a kibbutz. Rippin said, “A er a few weeks,
he quit. He felt that he couldn’t go back to his kibbutz in the uniform of a
jobnick.” A “jobnick” refers to a soldier in the Israeli military who doesn’t
directly fight. He might have a desk job in logis cs, or in intelligence, or
work in the IDF’s public rela ons unit – all are important jobs that make
the army and the country run – but they’re not seen as risking their lives
for their country. “The army will always put those who risk their lives on a
higher pla orm. The most extreme case is the fighter pilot. They are looked
at differently, and always will be,” according to Rippin.
Of the approximately seven hundred Talpiot graduates, several decided
to move into combat posi ons, deferring work in research and
development or other more tradi onal Talpiot post-gradua on jobs.
One of those soldiers came through Talpiot’s eleventh class. Because of
his role in the Talpiot program and IDF service, Israel’s Ministry of Defense
would not allow his name to be published. He would later become a
commander of Talpiot, but before that he worked his way into one of the
IDF’s most daring units, Shaldag. This small, special opera ons unit is
a ached to the air force. The soldiers in this unit are some mes dropped
deep behind enemy lines to carry out commando raids and other secret
missions. Though they are not pilots, they are responsible for one of the
most important tasks in the IAF.
It is believed that soldiers from this unit were sent to the Deir ez-Zor
region of Syria in the days and weeks before Syria’s nuclear reactor was
destroyed by Israeli warplanes. They were asked to collect soil samples to
make sure the site was what Israel’s intelligence community had suspected
it was.
They use lasers or electronic devices to hit a structure that will help
guide bombs or air-to-ground missiles to the exact correct spot. This is
especially useful for small and sensi ve targets. It’s also a way to limit
collateral damage when targets are in heavily populated areas, as the
enemy o en operates in populated areas. When Israel feels it has no
choice but to strike those targets in order to protect its own ci zens, the
desire to limit civilian casual es on the other side is always strong. World
opinion quickly turns against Israel when civilians are harmed.
That former member of Shaldag is far from being the only Talpiot who
morphed into a true commando. Several other Talpiot cadets went to
special officers’ school right a er gradua ng and became platoon leaders
before going on to join the figh ng in Lebanon.
The remarkable story of one of those fighters (whose current status and
sensi ve security posi on necessitates withholding his name) con nues to
inspire recruits. I will call him Natan. Friends and soldiers under his
command say he’s “the salt of the earth.” And he looks like the warrior he
always wanted to be. His hair is cropped short to the skull. Built like a
middleweight champ, his shoulders are big and broad. Not the type of guy
you’d dare into a fis ight.
Natan is from a small agricultural town a few miles south of Lebanon. A
few miles to the east is the Syrian border. As a child, he prac ced safety
drills, and he remembers the rocket fire from Lebanon being a li le too
close for comfort.
Of the genera on called “the children of the Yom Kippur War,” he was
born shortly a er the figh ng came to an end. At the age of twelve, a
doctor told him he’d never reach his dream of becoming a fighter pilot
because he had far from perfect eyesight. His mother was terribly
disappointed for him, but he happily told her, “Instead of flying planes, I
will build planes that don’t need pilots.” From that point on, without
knowing much about Talpiot, he was guiding himself in that direc on.
Because he feared recruiters would not find him in his small town, Natan
decided to apply to Talpiot, a rarity in those days. As he made his way
through the labyrinthine applica on process, he was in midated by the
“room full of generals” throwing complex ques ons at him. But it was a
kinder, gentler Colonel Avi Poleg who led the interview. The skinny small-
town kid, who would later go on to become an important fighter in the IDF,
was asked how a microwave works. “I remember having almost no idea,
but making up a coherent answer. The rest is pre y much a complete
blackout. I was very, very focused, and tried to keep my head up and not
look down. I came out with almost no recollec on of what happened.” He
also thought his chances of being accepted were zero.
As Natan started thinking of alterna ves, he heard that more than half
of his high school class had been accepted into the paratroopers. Resigned
that his weak eyesight would preclude him from going into a combat unit,
he thought of “doing something else,” when the phone rang. He’d been
accepted to Talpiot. He was ecsta c!
His Talpiot class was assigned to basic training with the paratroopers, a
unit he’d always admired. While not the top student in his class, when it
came to military life, he excelled. This cadet thrived on visits to army
outposts, naval ships, air force bases, ar llery teams in the field, armored
units, and research and development teams working on exci ng, futuris c
projects. “My favorite part of Talpiot was those special sessions when we
visited and got training between semesters. I remember coming back
home and seeing all my friends who actually went to combat units. I could
say ‘last month I did your training, this month I did yours.’”
The students in his class were known to be excep onally eager to
ques on the upper echelon of the Israeli high command.
One day, the class toured an air force base and heard a lecture from
General Avihu Ben-Nun, commander of the Israeli Air Force from 1987 to
1992. As a pilot, Ben-Nun became an ace, with at least three confirmed air-
to-air combat kills – two Egyp an MiGs and a Russian MiG flying for Egypt.
He was revered throughout the country.
The general was explaining his decision to buy more F15s than F16s. The
Talpiot class he was addressing had been in the army for less than two
years. Natan recalls, “One of the guys raised his hand and actually
challenged him: ‘That was a terrible decision. How could you do that?
Can’t you see what the air force really needs?’ In the end, Ben-Nun
explained why he did what he did, and he clearly won the argument. But in
how many countries do twenty-year-olds ques on the wisdom of
legends?” The student-turned-warrior con nued with a mischievous smile.
“When we met big commanders, it took us no me to ask ques ons. That
was our spirit. Everything was challengeable. I don’t know if it was
annoying to the commanders and embarrassing to our officers, but as I’m
older now, I can see how it would be.”
As this excep onal young cadet hit the middle of his second year, he
began to have second thoughts about the academic track of Talpiot. He
loved being in the field and felt that he might be able to contribute more
there. He started toying with the idea of going to field officers’ school, with
hopes of leading a platoon. His commanders told him they could help him
move in that direc on, but they urged him to finish his classwork and
degree first.
Despite enlis ng in Talpiot, Natan never lost his desire to become a true
warrior. A er gradua ng, his classmates took their assignments – mostly in
research and development – while he hit the dirt, literally. He moved back
into basic training, then worked his way into officers’ school and
commanders’ courses.
His aim was for a unit that un l recently had been a state secret, an an -
tank missile unit. The missile, called “Tamuz,” is a shoulder-fired guided
missile developed by Rafael. It can also be fired from a mounted posi on
on a jeep or small armored vehicle. It is propelled with a small amount of
solid fuel, making reloading rela vely easy.
The Tamuz has been used several mes during the Syrian civil war by
Israeli troops patrolling the border. When Syrian mortar shells are fired into
Israel by the Syrian army or by rebels, on purpose or accidentally, Israel has
o en responded by destroying the source of that ground fire with the
Tamuz missile.
Natan clearly has enjoyed his service in the field and is proud of it.
Enthusias c about the Tamuz, he says, “The missile is guided by a
television control. You have a camera at the head of the missile and you
just maneuver the missile to the target. For me, it was a great combina on
of state-of-the-art technology and a field-focused unit.”
For more than a dozen years, he was in that same an -tank unit; a
walking, talking, tank killer. If war came, his objec ve would be to use the
element of surprise to take out enemy tanks without the IDF having to
deploy full tank units in those areas. Among other things, his unit is trained
to open fire on tanks several miles away without being detected or tracked
by the enemy.
He’s now a lieutenant colonel in the reserves, in charge of up to four
hundred men, and he trains his teams to be capable of reloading and firing
in fi een-second intervals. As a reserve officer, he is an anomaly for many
reasons, one of which is that he sincerely enjoys his reserve duty. For him,
the ideology of service is mandatory. He believes “The reserves are the real
green army and need to stay in shape and stay ready. Our reserves have
fought for and saved the country in all of Israel’s wars. I found my home
here in the reserves.” He serves seventy to eighty days a year, a very high
rate for Israelis. Most serve in the reserves for a few weeks a year, at most.
Military historians agree with his assessment of Israel’s reserves.
Because the standing army only has approximately 175,000 men and
women, the reserves are especially crucial. There are almost four mes as
many reserves as there are regular soldiers. In the past, the main goal of
Israel’s standing army, when under a ack, was to hold off the enemy un l
the reserves could get in place. While warfare has changed, as has Israel’s
army, generals and poli cal leaders know that surprises happen and the
reserves are s ll held in very high regard throughout the country.
This modest Israeli from a small town, who feared he’d have a less than
stellar military career, is o en asked by Talpiot to come back and lecture to
the recruits. Of course, he always says yes. Natan has been instrumental in
inspiring more Talpiot graduates to move into the ground forces, and the
IDF hopes that s ll more will follow in his large, capable and dedicated
footsteps.
CHAPTER 15
ISRAEL’S NEW HEROES

IThey
srael loves America. They look to the USA as the land of opportunity.
see it as a great place to shop, so much so that Israelis visi ng the US
bring an empty suitcase for the new clothes they’ll buy and bring home.
Israel also eats up American television. They love Seinfeld, The Simpsons,
Sex and the City – the list goes on and on. One day Assaf Harel, an actor
and writer in Israel, was watching an episode of the HBO original series
Entourage. If you’re not familiar with it, the show is about a kid from
Queens who makes it huge in Hollywood, becoming a big star. His older
step-brother was already a b-list actor, but he also brings his two closest
friends to California with him. One becomes his manager, the other runs
the homes where the four live together in some of southern California’s
nicest neighborhoods.
Harel was watching reruns of the show when news came in that
Mirabilis, the company behind the instant messaging computer program
known as ICQ, had been bought by AOL for $287 million and another $120
million in deferred payments. It was the most anyone had ever paid for an
Israeli company, by far. Israelis were absolutely cap vated, and they quickly
learned the background of this incredible company and the reason for its
great market value. Mirabilis was founded by five Israelis in 1996. Four
were friends, the fi h was Yossi Vardi (now a legendary investor in Israel
and the father of one of the original four). They decided to come up with
instant messaging, a er realizing that the technology existed but that
nobody was even really experimen ng with it. Their goal was to be able to
connect computer users using Microso ’s Windows opera ng system.
Harel says, “When Mirabilis was sold it was a substan al event for the
young people of this country, in fact for everyone in Israel. A few guys
created something on their own and sold it for hundreds of millions. This
had never happened before in Israel. We had never had young millionaires,
especially ones crea ng their own wealth. I remember how enthusias c
everyone in the country was about it. And I thought if all of Israel was so
interested in reading about it, they would likely watch a show about it.
Since then, this kind of entrepreneurship has happened over and over. But
we were the first to really turn it into a popular culture event.”
Harel took the story line of the young Israeli millionaires and applied the
success factors of the American series. “Entourage was a dude series.
Dudes walking and talking. ICQ gave us the context for Entourage, Israeli
style.” He and his friends began wri ng immediately a er ge ng the idea.
A few months later, the show was given the green light and they started
shoo ng in 2005. Harel is the idea man for the show, but also one of the
stars portraying Guy Fogel, the most serious of the four main characters.
The show is called Mesudarim. In English, that tle would translate to
something like “se led for life.” The show is a drama c comedy about how
the four friends live their lives a er selling their high-tech company to an
American firm for $217 million. The four friends buy a mansion together.
They’re in each other’s business over women, over money and over how
they can possibly take the next step together in business. Mesudarim
quickly became Israel’s highest rated comedy series.
Harel is a smart and savvy producer, director, writer and actor. He knows
the country very well, and that helped him capture the essence of how
Israel was changing. He saw that it was becoming a society that so values
entrepreneurship and technology that he could turn a mul -million-dollar
technology sale into a TV series. And he astutely realized that the country’s
heroes had shi ed.
“It used to be that Sayeret Matkal (Israel’s Special Forces) was the unit
that everyone strived to be in,” he explains. “They were our heroes. They
were the stars: people like Ariel Sharon from Unit 101 in the paratroopers,
Ehud Barak and Benjamin Netanyahu, who were both from Sayeret Matkal.
They were our special clubs. But now it’s the high-tech units that are most
revered – 8200, and especially Talpiot. Israelis recognize high-tech players
the way Americans recognize athletes and celebri es.”
Harel believes the changes in Israeli society were born from the
disillusion of the Yom Kippur War in 1973. “Years ago, fighters created an
iden ty for Israel. Everyone went into army service and it used to tell you
who you were going to be. But a er that war, the na on started to change.
We went from an emphasis on brawn to an emphasis on brains. Talpiot and
those other “thinking” units represent Israel’s global contribu on to
innova on; brains are the new passport to global society. So now these
8200s and Talpiot graduates are our new heroes.”
The ba le between brains and brawn, individualism and the common
good, soldier and execu ve, has existed in Israel from the very beginning.
For decades, that ba le manifested itself in the rivalry between Yitzhak
Rabin and Shimon Peres. Many in the world always saw the two Israeli
poli cal giants as allies. But everyone in Israel knew otherwise.
“Rabin and Peres hated each other. Rabin was the fighter, the general.
Peres was never a soldier, but he was one of Israel’s first execu ves in the
Ministry of Defense. Fighters were taught to not trust people who didn’t
fight, and Peres resented that a tude. There was this everlas ng tension
between their two different approaches, and they were always up against
each other in the Labor party. It was uniforms versus suits.”
Assaf Harel sees Talpiot’s success in both in the IDF and in Israeli culture
as the ul mate win for Peres. “Un l recently, Israel was s ll in the era of
the fighter,” he affirms. “Now we’re moving on.” Assaf also believes it
won’t be long before the soldiers who were in elite technology units like
Talpiot will take on poli cal roles in the country, perhaps even the office of
prime minister.
Talpiot’s success is also evidence of Israel’s rota on from a socialist
society to a much more capitalis c society. As one who watches Israeli
society and culture closely, Assaf’s take is that money is now more
important than ever in Israel. “But Israel is far from the only country where
money is more important than it once was. That’s certainly true in the
United States and the Western world. That’s just the way things are.
There’s no use in judging it; you just do your best. You could argue that this
development is good because it promotes educa on. People strive to be
be er educated, they earn higher wages, and dona ons for charitable
causes come from that money.” That being said, his next project for Israeli
TV is the opposite of Mesudarim. It’s about old men si ng in a café all day,
griping about their lives.
With the changes in societal a tudes comes new popularity for Talpiot
recruits. Many members of Talpiot say that a er they enlist in the unit, it’s
amazing how many teenagers from their old neighborhoods and high
schools take the me to find their phone numbers. They call them to ask
about what they’re doing and about their path to Talpiot.
Saar Cohen is from Hadera, just north of Tel Aviv and he’s the first
student from Hadera High School sent to Talpiot. “Young kids track me
down” says Cohen with dismay. “They ask what will happen to me in
Talpiot? How do I get in?”
Obviously, they’re on to something. They want not only the educa on,
the training and the pres ge of the program. They want the added benefit
of being able to say for the rest of their lives that they are part of the
Talpiot elite. The rewards are many.
CHAPTER 16
“TALPIOTS ONLY NEED APPLY”

N etworking is a huge part of the Talpiot experience – for cadets, for


graduates in a full me army unit, and especially for those who have
moved on to the private workforce. Known casually as “Talpiots,” grads
hire each other, help each other find jobs and provide a boost whenever
possible. A er all, it’s a network of people born and bred to solve
problems, answer ques ons and unravel life’s riddles.
Marina Gandlin will one day soon take part in that Talpiot tradi on. For
now, she’s helping her boyfriend, an alumnus of Unit 8200, look for a start
to his career. “We were looking online for jobs for him and we were
shocked to see how many of these top jobs in communica ons and
computers say ‘Talpiot only,’” she exclaims. “For me, it will help. If your
company is looking for a leader, you might take an air force leader or a
combat leader. But if you’re looking for a leader and a tech specialist, you
might well say ‘Talpiot only.’ It’s the best of the brands.”
Those “Talpiot only” ads are o en published and posted on job sites by
non-Talpiot graduates eager to have Talpiots work at their company. But
when the company is run by a Talpiot, especially a start-up, the desire to
work only with other Talpiot graduates intensifies. Many would say they
speak a special language, they truly understand each other, there is a trust
from one Talpiot to another, they know how to get things done and they
share unique common experiences.
“Talpiot is a great pla orm for networking because it is such an in mate
program,” says Elad Ferber, of the twenty-fi h class of Talpiot. “Three years
and a few months of eighteen hours a day with the same thirty people, you
get to know the boys and girls very well. You literally can tell who they are
when you see their shadows. They’ll do anything for you. We feel a real
obliga on to help each other. This is especially true for anyone in your class
and for one year above and one year below you. This strong connec on
lasts during the program, into service and into the private sector. Nothing is
off limits. It’s a very ght knit community, and while we all need to make a
living, money never pops into the equa on.”
A few of the graduates of this elite club came up with a program called
Talpinet. It is an online forum for graduates only. It helps put one grad in
touch with another if they’re looking for someone to fill a certain role on
an execu ve team, if they need a programmer with certain skills, or if
they’re just looking to solve a problem previously thought unsolvable.
From Talpinet, air force pilot Arik Czerniak came up with the idea for
“Talpimeet.” Several mes a year, Czerniak finds a venue and invites all
seven hundred-plus of the Talpiot graduates to a one-night forum. At each
forum there are several speakers, Talpiot graduates who are rising to the
top of their fields.
One such forum was held in the spring of 2012 inside a large lecture hall
at Tel Aviv University. Graduates from thirty different classes showed up;
a endance was in the hundreds. They reunited, reminisced and heard
presenta ons by fellow Talpiots on how a cell works, grows, modifies, can
be mutated and cured.
They also heard from Elad Ferber, quoted above. Before he was
accepted to Talpiot, Ferber had been dead set on becoming a fighter pilot.
He’d already started the recruitment process. But like Arik Czerniak several
years before him, he decided that pursuing Talpiot would be a good thing
to do as well, so he began the parallel process of applying to both. One of
his oddest memories was being asked by the Talpiot interviewers to
iden fy Janis Joplin and Florence Nigh ngale, and to explain how World
War I began. “I knew Janis Joplin,” he recalls, “but I didn’t know exactly
when World War I started or who Florence Nigh ngale was – but I do now.”
Upon finishing his Talpiot studies at Hebrew University, Ferber worked
managing programs at the defense contractor Rafael. He believes Rafael
was very interested in him for several reasons, including his access to the
Talpiot network. “They know that when they bring one of us onto a
project, they have access to many more. They know we’ll use the Talpiot
network. I could a ract other Talpiots and access their ideas, even if they
don’t work on the same projects directly. It’s like a beacon for good ideas.”
A er his stay at Rafael, Ferber moved to the Ministry of Defense where
he managed large projects including some that are s ll very important to
the security of the country. His work had to do with a major so ware
upgrade. That project has been deployed and is currently being used by
the Israeli army. The rest is classified.
Ferber says it was those types of large projects – where he worked with
many people from numerous different backgrounds, represen ng many
interests – that helped prepare him for his future in business. As he
finished up his defense work for Israel, he began looking for his first real
step in the private world. Passing up a few opportuni es presented to him
by older Talpiot graduates, Ferber accepted an offer from Stanford
University’s MBA program. Like most Talpiots who go abroad to study or
start a business, Ferber vows to return home.
He sees similari es between his post-graduate program and Talpiot.
“Stanford’s MBA program is also ght knit and very intense. As in Talpiot,
we are greatly encouraged to work together; interpersonal rela ons are
very important. This gives me a second global network.”
While studying, he founded Echolabs. The company has devised a way to
test blood in a non-invasive way by using electro-op cs. Ferber says,
“There are a lot of consumer uses for this technology. It is less for the
medical community and more for athletes – it tells them how their bodies
are behaving under certain condi ons, when they should rest, when they
should eat and drink. It helps them op mize their bodies.”
Ferber constructed the prototype device himself, buying equipment at
hardware stores throughout the San Francisco area and pu ng them on a
wearable device the size of a wristwatch. The Talpiot network has been in
touch, helping him search for addi onal funding.
Rotem Eldar was in the sixteenth class of Talpiot and graduated in 1994,
nine years before Ferber. Eldar works for Gemini Venture Capital, one of
Israel’s largest VCs. Eldar was immediately interested in Echolabs. While it’s
not yet certain if Gemini will make an early stage investment, Eldar has
been helpful in pu ng Ferber in touch with the right advisors who might
be able to help with funding now and in the future.
Eldar works on the eleventh floor of a building with a commanding view
of the Mediterranean Sea in Herzliyah Pituach, one of the wealthiest towns
in all of Israel. There are sushi restaurants on the streets below and an
Xbox video-game system in Gemini’s lobby. He began working at Gemini in
2011, a er helping develop and market a communica ons company based
in Boston. He says the biggest part of his job at Gemini is due diligence,
making sure the companies he might recommend for investment would be
a good move. He uses the Talpiot network extensively to do this.
Poin ng out the window, he offers, “We call this Silicon Wadi. Fi y
percent of the startups are in a two-or three-mile radius of where we’re
standing. Microso is a block away. Broadcom is in that building over
there. In general, Israel and especially high tech in Israel is like a swamp.
Most people know everyone else. But it’s my Talpiot network that really
helps me get through the clu er and get reliable informa on faster.
“I know people from Talpiot who have been all over the place in
different areas of the military, and their paths have led to different
industries and different companies. That network is a huge advantage and
me saver to me and to Gemini. From where I stand, you can’t tell much
from the books or the financials of a company, especially if it’s a start-up.
What you need is informa on, and my Talpiot connec ons get that for
me.”
Moreover, once Eldar decides that it would be good for Gemini to pour
money into a certain project, having Talpiot connec ons can get Eldar and
Gemini in the door. “If it’s a good opportunity, it’s compe ve and difficult.
Many VC firms want to invest in good ideas – the next Facebook, if you can
recognize it – and everyone wants in. Personal rela onships are key. If you
know the entrepreneur or someone who knows him or her, you have a
be er chance of ge ng in, and ge ng in first. Let’s be honest; there aren’t
that many great companies to invest in.”
When a Talpiot is involved on the other side of the deal, Eldar says with
a smile, “We literally speak the same language; we’re on the same
wavelength; they feel comfortable speaking to me and that gives me an
advantage. Talpiot is the kind of network that breaks through boundaries
or ignores them.”
That sen ment is echoed by five Talpiot graduates who are all employed
by Takadu, an Israeli company that monitors urban water supplies and
systems. The owner of the firm, Talpiot grad Amir Peleg, brought fellow
Talpiots in soon a er star ng the firm. One of them, Haggai Scolnicov,
reflects, “If someone calls me up and the conversa on starts with, ‘You
don’t know me…my name is…I’m from Talpiot,’ it’s immediately a li le
different. They’re not all necessarily people I’d work with. But he probably
has something interes ng to say. True, there’s some element of the old
boys’ network, but it is generally very helpful. It’s a powerful thing. Where
else do you consistently have automa c contacts with people ten or fi een
years older or younger than you?”
The precedent for lifelong networking is set during their army years,
when Talpiots could ease each other through the bureaucracy. Some mes
poli cs, bureaucracy, or technical issues would slow down progress, but
Talpiots are able to push things forward faster working with Talpiots in
other units.
Uri Barkai, another Talpiot Takadu employee, explains, “Things get lost in
communica on. In order to talk to another unit, you have to go up the
ranks and then back down. So there can be some miscommunica on and
when you can cut through it by calling a fellow Talpiot, the informa on
flows faster, and that’s very good.”
Scolnicov recalls an instance when his commander wanted him to take
over a communica ons role. “I said, ‘I’m not so sure.’ Then I asked, ‘Who’s
on the other side?’ When my commander answered, ‘Simmy…he’s also a
Talpiot,’ I agreed immediately. Of course, the price a boss will have to pay is
that we’ll spend thirty minutes a day schmoozing.”
Biologist Ron Milo was never much of a cha er. Computer science,
physics, technology and chemistry are the tools he uses at the Weizmann
Ins tute, a world class research center in the city of Rehovot, south of Tel
Aviv, and just west of Jerusalem. Founded in 1934 by chemist Chaim
Weizmann, who would later become Israel’s first president, the Ins tute
was rated by The Scien st in 2011 as the top place for academics to do
research, outside of the United States.
It is a magnet for Talpiots like Milo. The nameplates on the office doors
indicate that it is filled with Talpiot graduates. At Weizmann, Milo and his
team are working on the “grand challenges of sustainability.” That means
they’re working on improving the efficiency of one of nature’s greatest
achievements – photosynthesis – to grow food faster where it is needed
most, in impoverished parts of the world where many people are starving.
He explains his complex research in a nutshell: “We are exploring the
possibili es, limits and op mality of carbon metabolism. We hope to
understand the fundamentals of its design principles, with the goal of
improving our ability to produce food and fuel more efficiently.”
Ron is very clear about what led him to this path at Weizmann, a perch
from which he literally is trying to save the world. “I am trained
scien fically and the training in Talpiot was deep and broad. It helped me
master math, physics and computer sciences, and that’s really key to who
I’ve become.”
While he has great respect for the program itself – from recruitment to
army drilling to instruc on to gradua on – he believes none of it holds a
candle to how the recruits drive each other, how the programs’ instructors
manage the cadets and how past Talpiot graduates impact the trajectory of
the careers (and lives) of those who come a er them.
“In Talpiot, who you are with and the people you’re exposed to is as
important, or even more so, than the training itself. You’re immersed with
interes ng and excellent people. You’re influenced by your commanders
and professors who constantly tell you that you can make a difference. By
the me you finish the program, you believe that and expect a great deal
of yourself as well.” Given that kind of heady environment, Milo believes it
is no wonder that once a Talpiot becomes involved in an important,
engaging project that he will recruit other Talpiots to join him. There’s a
natural affinity for similar pursuits. And once again, the network kicks in,
ensuring that fellow Talpiots coming on board can be trusted to prove
themselves valuable.
Rotem Eldar sums up the Talpiot advantage with these words of cau on.
“Remember there are no ckets to success in life. You can’t just rely on
saying ‘I’m from Talpiot’ and get whatever you want for the rest of your
life. But you do have the benefit of bonding with great people and
networking with great people. While you’re ul mately on your own, you
have a sort of built-in, personal advisory staff ready to help you find
answers and make decisions.”
CHAPTER 17
PROJECT SUCCESS

Igraduates
n the last chapter, we saw how the unique Talpiot network placed many
of the program at the forefront of Israeli industry. It’s fascina ng
to track the lives of former Talpiots, watching how their training, educa on
and army experience pay off later in their lives. Talpiots have an impact
everywhere they go, and they influence the economy of Israel in
unprecedented ways.
In 1993, the Internet was in its infancy. It existed only in a few areas
where researchers were experimen ng with networks. In some cases,
individuals with computers hooked up to phone lines could communicate
with each other through a hub with a fax-like connec on. One literally
heard a fax-like dialing system before a connec on could move forward.
Even to people in the high-tech world, the prospect that this small
network could one day be the center of global business, banking and
marke ng was s ll pure fantasy. But a small team of Israelis saw deep into
the future. They saw the value of the Internet; they imagined a world
where commercial and banking transac ons would be done online as a
ma er of course. They were way ahead of the game.
Marius Nacht, of Talpiot’s second class (whom we met in chapter 6), was
working for Optrotech, one of Israel’s first companies to be listed on the
Nasdaq, when one of the top managers there was allowed to create a
separate unit known as GAD. Nacht was invited to work for the new
division. The boss was very vague about the details, but Nacht took a leap
of faith. “He wouldn’t tell me what the project was about because he
didn’t want to spill the beans,” he explains. “But he was brilliant, so I
accepted the job offer without knowing what I was going to be working on.
A er I signed the documents, he told me we would be working on a big,
new, sophis cated printer. I was so disappointed with myself: a year or two
earlier I was saving Israeli pilots in combat and now I’m working on a
printer. I felt so stupid. He wanted me to run the so ware part of the
business which was extremely sophis cated. He said he’d only have five
people working there; the rest were outsourced.”
Optrotech would later merge with Orbit and then became Orbotech,
which s ll trades on the Nasdaq. The company manufactures, markets and
services products for the electronics industry, including circuit boards. It
also specializes in a process called automated op cal inspec on where a
camera takes very detailed images of an electronics component and then
reports whatever problems may exist with the device.
Nacht admits knowing nothing about so ware back then. “When I was
in Talpiot, there was almost no computer science prac ce. I didn’t even
know DOS, and back then it was DOS only.” Somehow, he later taught
himself UNIX and Macro-S before learning how to program on Apple’s OSX,
all of which were needed at Optrotech.
Then Nacht ran into a problem trying to write code that would allow a
computer to operate the sophis cated laser printer he was working on. He
remembers, “Nobody could solve this problem. I was told to call this guy
named Gil Shwed. He had served in Unit 8200 and was working as a
freelancer for Sun Microsystems in Israel. He hated corporate life and
would only work on one project at a me before signing on for a new task.
He was a freelancer of freelancers. So I invited him to come help me, and a
few days later he showed up. He was si ng in front of the monitor and
started typing as I was describing the problem – not even what I wanted to
do – just the problem. I kept talking, and he kept working. I shouted, ‘Hey,
why don’t you listen to what I’m trying to tell you?’ He didn’t even look up;
just said ‘don’t worry, keep talking.’”
Nacht laughs. “As a programmer, Gil is amazing. Even today, nobody can
match him. In three hours he solved a problem we had been working on
for a month – and our team included a man who had a doctorate and
another who had a master’s degree in computer science.
“He walked out of Optrotech that a ernoon – ‘ciao, arrivederci.’ But we
stayed in touch. I convinced him to come to work for the company. This
was quite a feat. He never took orders from anyone. For him to have a boss
and a paycheck was absurd. I’m very proud of being able to convince him
to come work for me.”
Nacht and Shwed le Optrotech a few months later to work for a New
York-based company in Israel that automated warehouse inventory. Their
clients included Boeing and Proctor and Gamble. Their storage data was
outdated, it was running on a system called VAX and Shwed and Nacht
converted it all to UNIX.
According to Nacht, all this me Shwed would say “I have a great idea.
The Internet is the next big thing. It will need to be secured.” In Shwed’s
mind was the formula for what would later become known as a firewall,
so ware designed to protect a user from all kinds of outside computer-
based a acks.
Nacht admits, “I really had no idea what he was talking about. I didn’t
know anything about the Internet. Even the concept was hard for me to
grasp. But Gil convinced me to quit Optrotech with him saying, ‘This is
happening now; we have to do something.’ That’s how Check Point
started.”
Even though Check Point was mostly Shwed’s idea, Nacht says he
generously offered to split the company 50-50. Nacht understood he was
about to be buried in an avalanche of work. He took a few weeks off to go
to Jamaica to just think and relax in “paradise.” While he was on the island,
he got a call from Shwed asking if he would mind bringing on a third man,
Shlomo Kramer, a friend of Shwed’s from Unit 8200. Even though Nacht
didn’t know him directly, he agreed. The new split was a third for each
man.
While Gil was the main driver behind the new company, Nacht recalls,
“He didn’t even have a computer. So I gave him my apartment; I had an
Intel 386. My bedroom became our office. A short me later, Shlomo’s
grandmother died. Her apartment became our office and I got my
bedroom back.” Despite having some seed money, Shwed, Nacht and
Kramer agreed they didn’t want to hire anyone to help them un l they
were confident they could start ge ng customers and sustain themselves
without outside funding. A er all, if they were to fail, they didn’t want to
take anyone else down with them. It would have been bad for their
reputa ons in a small country, but even worse, it would have been “bad for
their souls.” So they did most of the small work, the dirty work, the
organiza onal and secretarial work by themselves.
Once things began to look a li le more secure, one of their first hires
was a woman named Limor Bakal. Her job was to help the trio organize
their thoughts and their schedules, providing whatever support she could.
She started up the sales and marke ng departments; at that point,
everyone did everything at Check Point.
Bakal says, “I didn’t even know what the Internet was. It was something
special at the me, but it was all very technical and academic. We had to
educate ourselves before we could approach a poten al customer. First
you had to explain that they could use the Internet to do business. Then
you could talk about poten al problems and our solu on. You didn’t say to
a poten al customer, ‘Connect and use a firewall.’ You had to say, ‘This is
the Internet.’ I remember saying to possible customers, ‘Do you have a
website?’ If they said no, I wouldn’t bother with them because they were
way behind. But in truth, for a tech company, we were way behind
ourselves. The first me they brought in a personal computer, none of us
wanted to touch it – we were all scared to death. They’d say something
about Microso and I would say, ‘What the hell is a Microso ?’”
Bakal recalls Marius Nacht chasing accounts in the United States. “He
was traveling all the me. He would work and sleep in his car for months,
just like a classic Jewish salesman of old. He’d literally go to companies and
knock on the door.”
When I told Nacht what Bakal had said, he replied with a smirk, “That
sounds like me.” On one of Nacht’s early calls to a company in the United
States, he met a chief technology officer near Boston and the guy said,
“There’s no need for Internet security, no need at all.” The company was
the so ware maker, Lotus. “They laughed me out of the office,” says Nacht.
They thought no protec on would be needed for the Internet, that the net
would never be a communica on tool for commerce. The first me we
went there, the network administrator told me, ‘I’m never going to connect
to the Internet. I won’t need a firewall.’ What can you say? There’s really
no need for a firewall if you’re not going to be connected to the Internet.
Eventually that guy was kicked out, and the person who replaced him
called us. I showed him the product and he was so excited he wanted to
deploy it immediately on the gateway to the lotus.com network. I told him,
‘Look, why don’t you try it on a produc on network first for a bit, learn the
rules and learn how to use it, because if you deploy a wrong set of rules on
the firewall you’re going to kill your own produc on.”
Another early hire at Check Point was Gil Dagon, a Talpiot graduate from
one of the program’s first classes. Legend has it that he was equal to a
whole team of programmers. Dagon was and s ll is wholly focused on the
things that interest him most. In fact, he gave away many of his early
op ons with Check Point, op ons worth millions of dollars. “Don’t worry,”
Bakal chuckles, “he s ll got plenty.”
From its humble beginnings, Check Point is now a $12 billion company
and one of Israel’s biggest, most successful and most respected companies.
It is s ll a global industry leader and the firm holds several key patents,
con nuing to shape and reshape the face of online security.
To find more Talpiot corporate success stories you don’t have to look
online. Talpiot graduates have used the Talpiot system for problem solving
in numerous other industries. Talpiot is famous for indoctrina ng its cadets
in the “systems approach.” Instead of taking a limited approach to
something, look at all the factors in the situa on from above and find a
way to understand the impact as a whole before devising a plan.
Like many before him, Gilad Almogy had envisioned himself as a combat
commander and had been very suspicious of Talpiot before becoming part
of class number five in 1984. Yet he was an excellent problem solver and
did well on tests, some mes without fully grasping all of the concepts.
“Maybe that made me perfect for Talpiot,” he quips.
Almogy was already a student of the systems approach even before he
became part of Talpiot. But he credits Talpiot with making him a systems
approach master. “The way Talpiot taught it gave me a great background.
Later, I made a real living doing that kind of thinking.”
He du fully finished his three years of Talpiot classes at Hebrew
University, specializing in physics and math, then it was off to combat
officers’ school. As a brigade platoon commander in Golani, he fought in
the first in fada and served for a me in southern Lebanon. While lying
cold in the dirt during maneuvers, he’d o en look around at his young
soldiers and think to himself, “I bet none of these guys have advanced
physics and math degrees.”
A er his military service, Almogy was hired by Orbot, where he
perfected a system that finds semiconductor defects completely invisible to
the naked eye, saving companies hundreds of millions of dollars. In order
to work, semiconductors must be constructed perfectly. There can be no
chips or scratches on the small devices. (That’s one reason why the people
who develop them wear suits similar to what you’d see at a biochemical
disaster site.)
“You’re looking for possible defects that are a few tenths of a millimeter
large,” explains Almogy. “It’s like being told to find a grain of salt in a
football field. By the way, there’s grass on the field; ignore that. And you
need to know what kind of a grain of salt it is, and you have to be 100
percent sure that it’s not pepper.”
When Orbot was purchased by Applied Materials (known by its Nasdaq
symbol, AMAT) for $110 million in 1996, Almogy moved over to AMAT. He
was constantly in demand, rising to the rank of senior vice president, a er
receiving a doctorate in applied physics from CalTech.
He always wanted to start his own company, but it was hard to escape
from the corporate giant. “I can look at a full problem, that’s my strength,”
he reflects. Ul mately, he took a long hard look at his own problem, and
found a way to break free from AMAT. He knew that mul -system
management would be the key to his success.
Almogy founded the California-based solar energy company Cogenera in
2009, and he was able to coax two other top AMAT execu ves to come
with him. He says making a solar panel is similar to building
semiconductors in that every panel must be flawless to achieve maximum
efficiency. The company’s new technology is taking the world by storm with
its unique design for solar panels, and it has been able to a ract some big-
name investors including Vinod Khosla, one of the most successful venture
capitalists in history.
Much of Cogenera’s success lies in what Almogy calls “the Ikea
approach.” All of the raw materials are shipped to his plants for assembly.
He doesn’t have sole suppliers for any of the materials; that way he never
has to put all of his eggs in one basket. His company has won several big
contracts in California, including at Facebook’s new headquarters, The
Sonoma Wine Company and the Clover Dairy, as well as two very large
customers in India.
Talpiot grads also stand at the forefront of several technologies s ll in
the development stage. High on the list are mobile technology and self-
driving cars. Talpiot grad Itay Gat, of Talpiot’s fi h class, is vice president of
produc on programs at Mobileye. While Israel was figh ng terrorism in
Gaza during the summer of 2014, Mobileye was becoming a public
company, lis ng on the Nasdaq under the cker MBLY. On its second day of
trading, a er its successful ini al public offering, the Jerusalem-based
company boasted a market cap of $7.6 billion.
Mobileye is almost “a third eye” for your car. It can tell if you’re about to
bump into the car in front of you, and it will even step on the brake, if the
driver reacts too slowly. Mobileye also warns of other collision threats on
the road, including other cars and pedestrians, by sounding an alert to
quickly mo vate the driver to take ac on.
Mobile strategy is another industry a rac ng Talpiots. A graduate of the
ninth class, Guy Levy-Yurista has degrees in electrical engineering from Tel
Aviv University, a PhD from the Weizmann Ins tute and an MBA from The
Wharton School at the University of Pennsylvania. He is the father of five
US patents that focus on encoding, detec ng unauthorized computer
programs and op cal pulses.
He has spent his life in research and development, first in the IDF and
then in the private sector. Levy-Yurista worked to develop mobile pla orms
for AOL and security so ware giant McAfee’s mobile protec on unit. He
was the chief technology officer at a company called AirPatrol, which
allows the people behind conferences and other business-to-business
pla orms to control how invited guests use their mobile devices when
around sensi ve and proprietary informa on. It also keeps uninvited
intruders from being able to access programs on their smart phones –
including the Internet, tex ng and cameras – while in the vicinity of the
conference.
Another Talpiot graduate has had so many corporate successes, his
friends nicknamed him “The Idea Machine.” And Ariel Maislos’s ideas
became fabulous moneymakers. A er gradua ng from Talpiot in 1994,
Maislos served in an elite research and development unit un l 2001. Since
leaving Talpiot, he has founded and sold two companies. The first was
Passave. The company’s mission is to make connec vity be er, faster, less
expensive and simply more efficient for homes and businesses using video,
voice and high speed Internet lines. In 2005, revenue was coming in at a
pace of $30 million a year. The company was bought by PMC-Sierra in 2006
for $300 million.
Ariel’s other giant success was Anobit, which was able to score sixty-five
global patents. He and his co-developers came up with a way to make flash
storage devices hold more informa on. While that technology would be
prized by any company in the technology sector, it is most valuable to the
mobile device market. A er being courted by several big name companies,
Anobit was purchased by Apple for $390 million in 2012. Maislos stayed on
at Apple for a short me a er the sale of his company, but le in a few
months later, presumably to start another new company that will sell for
hundreds of millions of dollars as well.
Of course, money isn’t the world’s only measure of success. The name of
one early Talpiot graduate has been forever enshrined by music-lovers
worldwide. A er gradua ng from Talpiot’s course-load at Hebrew
University, Meir Sha’ashua used algorithms to develop new radar systems
for the IDF. Figuring out a way to use algorithms to improve sound, he co-
founded Waves Audio Ltd. in 1992. Their products record, mix and master
audio, and are widely used in the music and film industries.
Sha’ashua’s company also came up with a way to ease the nerves of
millions of soccer fans all over the globe. During the World Cup in 2010,
fans at South Africa’s stadiums were blowing vuvuzelas, loud horns that
make a very dis nct buzzing sound. On the opening days of the
compe on, the BBC alone reportedly received several hundred
complaints, all begging the network to do something. FIFA, soccer’s
governing body, refused to ban vuvuzelas from the matches, so Waves
Audio stepped in to provide the solu on. They quickly developed and
offered a special plug-in to television networks who could use the product
to drown out the buzzing.
The company won a Technical Grammy award in 2011. On the red carpet
in Los Angeles, Sha’ashua shook off credit for his inven ve products,
graciously telling interviewers, “It’s a great honor! I can only think of the
Waves staff – all the employees who actually made this happen.”
CHAPTER 18
LIFE SAVERS

W e met Eli Mintz, of the fourth Talpiot class, in chapter 3, and the
trajectory of his career inspired all who came a er him. A er his discharge,
Eli went to France to study business at INSEAD (one of the largest graduate
business schools, with branches in different parts of the world). His wife
Liat, a bio-scien st, found a job working at the Pasteur Ins tute in Paris, a
non-profit founda on dedicated to studying biology, diseases, vaccina ons
and micro-organisms. “In the early nine es, the French were leading even
the Americans in researching the human genome,” Eli explains. “Of course,
the Americans quickly caught up and le the French behind. But at that
point, there was a lot of effort in France. We were really in the right place
at the right me.”
While at INSEAD, he was trying to figure out what to do with his
algorithm and business exper se when Liat suddenly had an idea. The two
would combine their knowledge to develop a computer that would make
genomic data mining faster, more reliable and more effec ve.
From that idea blossomed Compugen, one of the first companies to use
sophis cated algorithms to work on genomic data mining and mapping of
the human genome. True to form, he founded it along with a few other
Talpiot graduates, Simchon Faigler and Amir Natan. (Later they’d add
another Talpiot to the team: Mor Amitai would become a long me CEO of
Compugen and one of Talpiot’s brightest business success stories.)
Together they developed a computer that could map and analyze DNA,
enabling pharmaceu cal researchers at drug companies like Merck, Pfizer,
Bayer and Eli Lilly to search gene c codes in order to develop more
effec ve drugs.
But it took the marke ng know-how of an American to take Compugen
to the next level. Mar n Gerstel grew up in the United States without
paying much a en on to his Jewish heritage. He just wasn’t that into it; it
didn’t seem relevant as he went to Yale, then worked his way up the ladder
in California’s booming bio-tech industry. He made a real name for himself
as CEO of Alza Corpora on, a firm that makes drugs that fight everything
from AIDS to ADD.
While on a business trip several years into his career, he met a young
Israeli woman. She convinced him to visit Israel shortly a er mee ng and it
changed his life forever. He felt at home the instant he got off the airplane,
a feeling many non-Israeli Jews describe regarding their first trip to Israel.
But Gerstel did something about it. He married that young woman and
became a serial backer of great ideas that were popping up in Israel’s
business community. Among the companies that a racted Gerstel as a
financial backer and management consultant, was Compugen.
When Gerstel met Mintz and other Talpiot grads, he was extremely
impressed. “There’s nothing like Talpiot anywhere else,” he says with
enthusiasm. “The people that come out of the program think differently
than the typical people coming out of even the top universi es anywhere
in the world. Just look at how they got this way. The Talpiot program is
built on a desire to serve the world and your country. It’s such a
concentrated effort for nine years. Think about that, nine years when
you’re only eighteen. They’re learning, and they’re developing, and
applying their knowledge to real problems where the lives of their brother,
sister, cousin and parents may depend on whether they’re right or wrong.
You can always count on the technology here. Marke ng? They don’t have
a clue. But when it comes to technology, Talpiot teaches them to be the
best.”
Brought in by some of Compugen’s bigger investors to “business-fy” the
company, Gerstel steered the company through a major transi on shortly
a er Compugen listed and rose on the Nasdaq.
And he immediately realized that Compugen was vastly underpricing the
best product on the market. “Their computer was much be er than the
compe on that was selling for $1.2 million. Compugen sold its version for
$30,000.” Ironically, the high quality of Compugen’s computer caused a
drop in its sales. “They were so good and so fast nobody needed another
one,” Gerstel recalls.
He came to firmly believe that Compugen needed to rebrand itself from
a computer company to a life-sciences company. Yet his idea to change the
focus of the company would soon lead to the exodus of the company’s
founders.
The original Talpiot team also had a beef with conven onal wisdom in
corporate circles. As Gerstel led the four – he calls them “kids” – from
mee ng to mee ng with the biggest pharmaceu cal companies in the
world, it was apparent to him that they were right and everyone else was
wrong. “I quickly realized that these kids in the room, the Compugen guys,
knew more about biology and the building blocks of life than the rest of
the world. We visited with a number of major pharma companies who
would say, “You’re good mathema cians, but your theories can’t be true.”
They couldn’t get away from their central dogma of biology: one gene, one
transcript, one protein. So we came back and set up a bio lab to test the
predic ons we found in our computer. And we found they were correct.
Over a number of years, slowly but surely, we transformed ourselves into a
life-science company. We hired more biologists; we increased our
investment in lab work and research.”
In essence, Gerstel and the new wave of Compugen execu ves put the
company on a new track where it had a be er opportunity to a ract new
clients. And that second genera on of management was also provided by
Talpiot. Gerstel looks back with reverence. “It’s the way they think,” says
Gerstel. “The issue for the Talpiot graduate is to first iden fy the problem
and then get to the bo om of it.”
With signature energy, flair and pride, Gerstel says that Israel, (thanks in
large part to Talpiot and Compugen) is the world leader in algorithms.
“Nobody does it be er. And Israel should be a center for this. It has the
algorithms and the biology, both among the best. Three of the last seven
Nobel Prize winners in life sciences have been Israeli. They shouldn’t give
Nobels in Stockholm, they should give them in Jerusalem or Tel Aviv.”
Occasionally Gerstel has been accused of being too enamored with
Talpiot, to the point of embarrassing Compugen’s execu ves during sales
pitches. One former Compugen execu ve remembers being especially
embarrassed. He would say to his friends, “I travel all over the world and
nobody outside of Israel has ever asked me about Talpiot. Even inside Israel
it is very rare because people just don’t know about it. Mar n would give
everyone a glowing lecture about Talpiot, and I would have to just sit there,
while the pharma execs would stare. It was awful…but he is a good
salesman.”
While founder Eli Mintz and Mar n Gerstel didn’t get along well in
business, years a er their high profile Compugen split they only had praise
for each other. Mintz remembers successful seed investor Jonathan
Medved pu ng up money, “but his most important contribu on to
Compugen was the introduc on he gave us to Mar n.” Mintz con nues,
“We were so fortunate to meet him. He was really important in making
Compugen what it is. Having Mar n Gerstel on board was a huge plus
because of his business experience, his connec ons, his ability to raise
capital, his strategic sense and his understanding of the biotech community
through his experience at Alza. We were really entering into biotech
without anyone in the company having experience in biotech. He was a
great guide.”
When hearing of Mintz’s praise, Gerstel was genuinely touched: “I never
knew he felt that way.”
Just as serendipity brought Gerstel to Israel, a chance occurrence
changed the future of Guy Shinar, of Talpiot’s eleventh class. Before joining
Talpiot, his goal had been to eventually become an air force pilot or a naval
commando, and he had hoped to con nue in that direc on. But during
basic training, his eye was hit by shrapnel during a live fire exercise. Told
that his vision would be permanently impaired, he abruptly realized that
his dreams of combat service were finished.
But he would more than make up for it a er a long and difficult road. His
first year of Talpiot study at Hebrew University was excep onally
challenging, but eventually he fell in love with science. “I knew it was
something I wanted to do for the rest of my life.”
A er gradua ng he went on to research and development, in Israel’s
ordinance corps and then in the intelligence corps. Though he s ll felt that
he was missing out on combat while developing new systems through
research and development, he started to realize that the contribu on he
was making to Israel’s defense was important and would be long las ng.
“That’s life,” he reflects, “You choose one thing and give up on the others.
There are always pros and cons.”
A er leaving the army a decade a er star ng, it was off to business
school in France. At this point, Shinar was star ng to think more and more
about the growing medical device field in Israel. Immediately upon
returning to Israel, Shinar became the first person hired by X-Technologies,
a company that was formed to develop and sell catheter technology for
cardiac pa ents. It specialized in making angioplasty balloons used by
heart surgeons to widen obstructed arteries. A balloon is inserted into the
pa ent and inflated, allowing for be er blood flow.
Four years a er its founding, X-Technologies was bought by Indianapolis-
based Guidant for $60 million in cash, plus another $100 million for hi ng
set sales markers. Guidant reportedly agreed to aggressively market X-
Technologies’ product. Three years a er the close of the deal Guidant
became the buyout target of several major corporate tans including
Johnson and Johnson, Boston Scien fic and Abbo Labs. In the end, a bid
from Boston Scien fic aided by Abbo Labs won the long takeover fight.
Despite specific guarantees from Guidant promising to market and
promote X-Technologies’ products, Guidant instead focused on other ideas,
pu ng X-Technologies on the back burner. Shinar, X-Technologies’
founders and top investors sued Guidant for failing to properly market and
meet sales goals. In the end, the case was se led, leaving Shinar and his
fellow plain ffs short of the original $160 million sales figure, but s ll very
well off.
At the me, Guidant’s buyout of X-Technologies was one of the first
sizable purchases of an Israeli medical device company. That deal signaled
to the world that Israeli companies and technologies were on the global
business stage.
Shinar went on to serve as a board member of several other Israeli
medical device companies before co-founding a company called Javelin
Medical, where he also serves as chief technology officer. The goal of
Javelin is to prevent stroke in high risk popula ons, specifically pa ents
with atrial fibrilla on, a common form of arrhythmia (which involves an
irregular rhythm of the heart). Shinar explains, “There are no good
treatments for stroke right now, and preven on is the right strategy.”
Javelin is currently tes ng its technology on animals and hopes to move to
human trials shortly.
Like many Talpiot graduates, Shinar says he was drawn to the medical
device field because it requires mastery of various disciplines: technology,
medicine, clinical trial design, sta s cs, quality assurance, as well as
regulatory and intellectual property, to name a few. For Talpiots, it’s a
natural fit. “People who have gone through Talpiot have a genuine
compe ve advantage in working to solve several problems at once from a
variety of different perspec ves,” Shinar points out. “We were taught to
use the systems approach for years, and this is exactly the kind of skill set
required in the medical device industry.” Shinar credits Talpiot as being the
forma ve event in his life; nothing has had a bigger impact on him.
All that Talpiot crea vity and brain power was harnessed by an
enterprising Israeli business giant, Yossi Gross. Though his own military
service preceded the emergence of Talpiot, his entrepreneurship in
medical device technology has a racted a healthy cadre of Talpiot
graduates. It is a burgeoning field limited only by the scope of one’s
imagina on.
Gross’s personal history includes aerodynamics engineering and a s nt
as one of the leading engineers on the Israeli Air Force’s Lavi fighter jet. He
le that posi on because he felt the bureaucracy of a big project at a huge
company was draining him of his crea vity. Shortly a er that, his wife
complained that her electric razor wasn’t working properly. Gross went
beyond fixing it: he transformed it into a product that eventually became
the world’s leading brand –”The Lady Remington Smooth and Silky”
electric razor.
However, Gross quickly soured on the consumer electronics business.
“One month I was developing jets for the Israeli Air Force. The next, I was
making ladies’ shavers. I went from high tech to low tech. It was
frustra ng.”
He didn’t know it then, but he was about to get his chance to move into
a groundbreaking area of high tech and to work with some of the best
minds in Israel and the world. Soon a er separa ng from Remington, he
met another Israeli entrepreneur who had an idea to develop mini-pumps
to deliver pharmaceu cals. “I didn’t know anything about it at the me,
but I told him I could make this.” A short me later, he took the idea and
his designs to the Irish biotech company, Elan. They invested in the drug
pump, giving Gross seed money to develop his own ideas and companies in
the new world of biomedical engineering.
Yossi Gross now has six hundred patents to his name and has started
more than a dozen companies in biomedical engineering. He groups most
of those companies under the name Rainbow Medical, which also serves
as a funding arm for the companies and technologies it owns. One of the
companies under the Rainbow umbrella specializes in developing
minimally invasive implants designed to help hearts work be er. Another
one of his companies efficiently reduces fat with ultrasound, lessening the
need for liposuc on procedures which are expensive and require a long
recovery period.
Since the beginning, Gross has looked to Talpiot to staff the companies
under his control. In Israel’s medical device sector, Talpiot graduates have a
clear advantage because of their high proficiency in engineering and their
ability to incorporate new technologies and new ideas. They’re also highly
respected because they can understand and manage several different parts
of a project at one me – from the mechanics to the technology, to the
medical realm, to so ware produc on.
The CEO of one of Gross’s companies, Nano-Re na, is Raanan Geffen, a
graduate of the third class of Talpiot. The company is working on a ny
ar ficial re na to help pa ents who have lost, or are losing their eyesight,
to see well again. Right now, the main candidates for the ar ficial re na
are pa ents suffering from age-related macular degenera on. Trial tests
con nue as the product inches toward perfec on.
Geffen spent most of his career developing be er communica ons
technologies and naval systems for the IDF. A er more than two decades, it
was me for him to leave. Because he was a crea ve innovator and had
experience managing large, mul faceted projects, many doors opened to
him. Looking back, Geffen knows the values that influenced his choice. “I
didn’t waste a minute during my twenty-three years in the military, and I
didn’t want to waste my me in the private sector. That was for sure. The
work I’m doing here is very meaningful; helping humanity is very important
to me.”
He gives credit to Talpiot for ge ng him to a point in life where he could
run a company in a field that’s s ll in the start-up stage, a field hungry for
new and innova ve ideas. “Talpiot shaped who I am today: it taught me
how to be innova ve and to have the confidence to use that innova ve
drive.”
A California company called “Second Sight Medical Products” has proven
that Geffen is on the right track in a growing field. They have a product
similar to the ar ficial re na being developed by Nano-Re na, and it has
already been implanted in several pa ents. Geffen is watching Second
Sight’s progress very closely. “They are a compe tor, but we’re roo ng for
them. They’ve proven, as we did, that this technology works. It needs to be
updated and get be er, but it works.”
One of Geffen’s first hires at Nano-Re na was another Talpiot graduate,
Kobi Kaminitz of the sixteenth class. We met him in chapter 12, working on
cameras and electro-op cs for Israel’s satellites. Kaminitz notes that
despite the name Nano-Re na, the company isn’t quite dealing with nano-
technology. The components are incredibly small, but not small enough to
be considered nano-technology. U lizing his experience in the military, the
ar ficial re na he is developing with Geffen uses a lot of the same
technology employed on Israel’s fleet of intelligence space satellites. “Our
goal is to use a chip less than five millimeters large, like chips used in a
cellphone’s digital camera slot. On one side is the lens; on the other side is
a series of pulses that sends signals to the re na. It replicates photo
receptors, rods and cones in the human eye,” he explains.
Another crea ve company under Rainbow Medical’s umbrella is
Maxillent. It specializes in making minimally invasive dental implants; its
top product is an innova ve way to conduct a sinus li , a procedure which
will increase the amount of bone in the upper jaw area.
Gideon Fos ck is the CEO of Maxillent. Fos ck’s grandfather le Belarus
in 1939 as World War II and the Holocaust were just beginning. He
remembers his grandfather telling him of the awful pogroms he escaped in
Europe. His grandmother was from Poland and lost her en re family in the
Holocaust. His family heritage was just one of the reasons that Fos ck was
so eager to dedicate a decade of his life to Talpiot, the army and his
country.
Fos ck first heard of Talpiot while he was a high school student in Tel
Aviv – and he knew he wanted in. He made it into the tenth class of Talpiot
in 1988. A er gradua ng from the academic por on of the program,
mastering physics and engineering courses and receiving degrees in both
fields, he moved on to a military intelligence technology unit. He quickly
became a research and development leader and assisted a sec on head
while working on several advanced military projects that remain largely
classified today.
He was awarded one of Israel’s highest honors, the Israel Defense Prize
for working on mul disciplinary systems. While all the details of the project
haven’t been unveiled, Fos ck’s work had to do with advanced alert
systems designed to detect the offensive movement of enemy missiles and
ground forces. The system combines the use of computer science, physics
and electronics to provide the IDF with more warning than ever before of
the threat of enemy a acks. His work demonstrates exactly the goal of
Talpiot.
Maxillent is growing an average of 15 percent a year, and he credits
Talpiot and his intelligence experience for his execu ve skills. Fos ck says
both Talpiot and military intelligence taught him to think differently.
“Talpiot drills into you, systems approach, systems approach, systems
approach.”
Apparently that emphasis has never changed. A few years ago, Fos ck
went back to Talpiot for a reunion, and one of the Talpiot classes at that
me performed skits: the punch-line was always “systems approach.”
Fos ck recalls, “In one skit, several students were on the stage ac ng out a
drama c scene. They went to leave the room and the door got stuck. It
seemed jammed, so the first student pulled and banged on the door.
Another then tried to force it. A third student looked at the door from all
angles, up and down, and side to side, for several minutes. The other two
asked him what he was doing. ‘Systems approach,’ he answered
though ully. Then he unlocked the door and it opened easily. Trust me – it
was funny.”
Fos ck calls Given Imaging, an Israeli drug company, a great example of
a company using the systems approach. It is most famous for designing a
pill with a camera inside that takes pictures of a pa ent’s stomach before
exi ng the system. “The idea came from a team that worked on guided
missiles at Rafael. They know how to make stuff small. They know op cs,
then they came up with something new. It’s a classic example of the
systems approach, pu ng together all of the stuff that we already know
and using it for a new purpose.”
The late Steve Jobs – who started and ran Apple, of course – wasn’t in
the IDF and it’s likely that he never heard of Talpiot. But Fos ck says, “Jobs
may have been the best systems guy in the world. He could always see the
big picture, from user interfaces to patents to marke ng and public
rela ons. He redefined the music industry almost by himself. He could look
at a problem from a wide perspec ve and create new ideas on how to
solve it.”
Gideon Fos ck points out another common characteris c of Talpiots:
genuine admira on for each other’s ingenuity and achievements. “There
was a brilliant Talpiot graduate from the fi h class. He loved his job, loved
research and development, and he was a great help to me in my life and
career. He absolutely cherished moments when I figured out something
that he could not. We were once working on a project involving op cs, and
the engineers kept seeing streaks. They couldn’t discern why. I finally
figured out that it had to do with moisture. It was the happiest I’ve ever
seen him! And that’s a large part of what makes the Talpiot community so
successful – the desire to help one another and to collaborate, without
worrying about who gets the credit.”
CHAPTER 19
REUNION!

Iyoung
n the pages of this book, we have met dozens of youthful Talpiot cadets,
men and women whose lives revolved around their experiences in
Talpiot and in Israel’s military. Educated and confident, they were eager to
help Israel progress; and a er discharge they were op mis c about their
own career prospects. What happened to them? Where are they today?
Imagine a room filled with Talpiot graduates, a reunion of more than two
dozen classes. They are seasoned veterans now, and their life stories are
varied and colorful. Some will briefly allude to their experiences
immediately following their army service or inform us of their current
workplace; others will spin long tales of enterprise and adventure.

Talpiot Class 2
We met Opher in chapter 4, when he was tapped by the IDF
and the Ministry of Defense to lead Talpiot. It was the first
me a Talpiot graduate was selected for the posi on. Star ng
with Talpiot’s seventh class in 1985, he modeled the program
somewhat along the lines of successful Ivy League schools in America,
establishing a sense of tradi on. He also oversaw an increase in the
number of women recruited for Talpiot.
He currently lives in Belgium where he serves as a consultant to
European technology companies and teaches at Ghent University, though
he plans on returning to live in Israel. Ghent is in a largely Flemish-speaking
part of Belgium. His colleagues have pleaded with him to learn the
language, but he playfully replies “I already speak one language [Hebrew]
of fewer than ten million people; I’d be crazy to learn another.”

Talpiot Class 2
Opher was one of the first Talpiot graduates to be given
intelligence field assignments. A er helping arm the forces in
Sinai with sophis cated long-range intelligence devices (1982),
he spent me at RCA in Camden, New Jersey. At the me, the
company had contracts with the Israel Defense Forces. When he returned
to Israel, he was once again a pioneer as one of the first Talpiot graduates
to lead new classes of Talpiot cadets. Like his classmate, Opher Yaron, he
set a successful example, proving to the army that it was be er to use
Talpiot graduates as commanding officers of Talpiot cadets.

Talpiot Class 2
When Boaz became a member of the second class of Talpiot,
the program was s ll secret and experimental. A er his army
service, he went on to a successful career in private industry
working in telecommunica ons. He did a lot of work with
Asymmetric Digital Subscriber Lines (ASDL) and other kinds of
communica ons that u lize greater bandwidth to move informa on
quickly.

Talpiot Class 3
A true global adventurer, Gilad has a spot-on nose for
business. Not your typical Talpiot, he confides that he was
probably one of its most reluctant recruits in the history of the
program. When he was being considered for Talpiot, his father,
an engineer, told him it was a waste of me: “What the hell are you going
to do with physics and math? There’s no profession there.” Moreover,
when Talpiot recruiters told him they would “teach him to think,” he
laughed at them. Yet by the me Gilad finished his three years of study
under Talpiot, he had much greater respect for its methods and goals.
When we last met Gilad in chapter 10, he had joined the navy and
became one of Talpiot’s first combat officers serving on a missile ship. In
his research and development career in the IDF, the bulk of his me was
devoted to developing an -ship radars that helped Israel deceive enemy
detec on systems.
Gilad used his experience in the navy and Talpiot to have one of the
most diverse and dangerous careers of anybody to ever graduate from the
program. A er spending me working on projects in Japan, he moved back
to Israel where he fell in love with a woman running Portugal’s commercial
office in Israel. At some point, the incessant threat of war and terrorism in
Israel was too much for her, and she wanted to move their family of four
children to Portugal. Gilad always loved to travel and to experience new
cultures, so off they went.
It was a mistake. He found Portugal to be an absolute mess, with an
unfriendly business climate and almost no entrepreneurial spirit, quite the
opposite of Israel. He had ideas, but there were so many roadblocks to
inves ng or star ng a company, he gave up. He reports, “In Portugal you
can’t fire anyone. You can’t create or innovate. Almost nobody in the
country really wants to work. You can’t run a country that way and hope to
move it forward; it will inevitably fall behind. The economic crisis that
exploded in 2009 is evidence of this.”
His cket out of Portugal was his father-in-law’s business connec on in
Africa. The awful, brutal civil war that had ravaged Angola for almost three
decades had just come to an end. Angola had been a Portuguese colony for
hundreds of years. For be er or for worse, those es created ripe business
opportuni es.
Gilad’s father-in-law wanted to sell Israeli technology to Angola, a
country that had very li le physical infrastructure and almost no
technology infrastructure at all. The Angolans wanted to catch up to the
modern world: they needed telephone lines, cellular networks, satellite
technology and the Internet. These are all areas where Israel excels.
So in 2004, Gilad moved his family to Cape Town, South Africa. Angola
was s ll far too dangerous a place to move his family. “I was adventurous
and hungry, not naïve and stupid,” he quips. A er se ng up his wife and
kids in Cape Town, he began making regular trips seventeen hundred miles
north to Luanda, Angola. There are several dozen flights a week, making
the trip rela vely easy, even by western standards. Even though the
commute was tolerable, Gilad says, “It was like living in heaven, but
working in hell.”
One day, he was driving a jeep with his wife outside of Cape Town in
what’s known as the bush, the unpaved countryside. They hit a massive
bump. “We were three hours outside of town. My wife started screaming
at me, but then stopped when she realized I wasn’t in the car next to her
anymore. I had flipped out head first.” He fractured his collarbone and
broke ten ribs. It took thirteen hours to get to the hospital. By then, his le
lung had collapsed, he suffered a concussion and he was in and out of
consciousness. “I was in intensive care for ten days, then laid up in the
hospital for three weeks. In order to reinflate a lung, you need to cough a
lot. But I had business to do in Angola! I asked the doctor what else I could
do. She said, ‘Run up and down the stairs over and over.’ Everyone in the
hospital was soon talking about that crazy Israeli running up and down the
stairs with an IV in his arm. But it worked. The doctor said she never saw
such a fast recupera on from a collapsed lung. Talpiot and the Israeli army
training I had been through taught me the dedica on to make this kind of
thing happen.”
Gilad admits to being a risk taker and says someone who plays it safe
would never try to do business in Angola. Crime and security remain
problems, and you never know exactly who you are dealing with. “You
have to have almost a sixth sense,” he winks. Accommoda ons for business
travelers in Angola are far from ideal. “Listen, I’d slept in the dust in Israel
for free,” he explains. “Certainly I’d sleep in the African dirt – with roaches
and Lord knows what else – for millions of dollars. I can’t say I loved being
there, but the discomfort wasn’t enough to stop me.”
Industriously, he imported and sold Internet capacity and storage
solu ons. “Growth was in the double digits,” he recalls. “But there was a
lot of corrup on and a lot of misery there. It’s heartbreaking. In the end, I
was a middleman selling everything from copper wire to network
appliances to data storage systems. I did not like it, not even for a moment,
but I never really felt that I was in danger. I happily le Angola for good.”
Working under nearly impossible condi ons prepared him for another
tough money mission several years later. While Gilad would neither
confirm nor deny it, he reportedly was inches from securing a deal with
Muammar al-Gaddafi to build resorts in Libya, a short me a er Libya
reached an agreement with the West to give up its weapons of mass
destruc on. The deal never quite materialized, and the Libyan leader was
violently deposed and killed in the “Arab Spring” of 2011.
A er working briefly for the Israeli defense firm Elbit, he a ended the
French business school, INSEAD. That prepared him to do what he is most
suited for: Gilad is now a matchmaker for private and corporate investors
looking to pair up with good ideas and good companies. He calls himself “a
cross between an investment banker and a scout. I can look at material,
analyze it and quickly decide if there is quality and value there. That’s what
I’m best at.” He works for himself – but his Talpiot-and INSEAD-filled
address book is what brings him business, new ideas and new connec ons.
He never pitches a deal that he wouldn’t personally invest in. “In my
business, your name is everything.”
Gilad has also joined with several other Talpiot grads to form OTM
Technologies, a company at the forefront of making devices that allow a
user to write notes or messages by hand on mobile phones and pads. His
device is called “the feather.”

Talpiot Class 5
Amir’s work was primarily in Israel’s burgeoning unmanned
aerial vehicle (UAV) program in the late 1980s and early ’90s.
Israel is now a world leader in UAV and UAV component
exports. His work with the program through Talpiot laid the
groundwork for that.
He went on to become a serial entrepreneur, founding three firms
including YaData, which he sold to Microso . He founded and is currently
the CEO of a water security company named Takadu (described in chapter
16), based in the industrial town of Yehud. The firm has clients all over the
world including municipal water systems in London, Chile and Israel.

Talpiot Class 6
An Israeli Ministry of Defense legend, Eviatar was later known
as “the right hand of Talpiot.” He was the founder of
Nachshon, a high school program that has furnished Talpiot
with several cadets. Currently, Eviatar is Israel’s top
commander on cyber-defense, repor ng directly to Prime Minister
Benjamin Netanyahu.

Talpiot Class 6
We met Tzvika in chapter 11 as the gutsy young man working
for the Israeli Air Force inside Elisra, a defense company where
he worked with engineers many years his senior. A er leaving
the service, he went on to work for two successful start-ups
founded by Talpiot grads. He currently works at Amir Peleg’s
Takadu, along with other Talpiot graduates Haggai Scolnicov,
Barak Peleg and Uri Barkai. His work involves designing and
tes ng the system that tracks urban water flow.

Talpiot Class 10
As a youngster living in Japan, Matan developed revolu onary
so ware that enables instant financial transac ons from one
end of the world to the other. A er Talpiot and army service,
he took a different route than most graduates. He went to
Hollywood.
Matan uses the mathema cal so ware development skills he learned in
an elite military intelligence unit to produce a website called audish.com,
essen ally an online cas ng agent. Its goal is simple: match directors and
film producers with actors and actresses online, elimina ng some of the
middlemen. When Matan isn’t developing the site, he is the managing
director of a bou que angel capital venture firm called Secent LLC in Santa
Monica. Secent describes itself as providing a “global gateway to early-
stage companies who have innova ve technologies with great poten al to
generate revenue worldwide.” In other words, he says, “we like to help
people make their dreams reali es, and we know people who know
people.”

Talpiot Class 13
Ophir found his niche in Unit 8200, developing so ware to
retrieve data stored in the Israeli military machines’ computer
servers. A er 8200, he founded a company called CloudShare
that did similar things for major corporate customers all over
the world. He explains that the servers are based in Miami, “because, God
forbid, if they’d be based in Israel, you’d never be able to a ract clients.
What could you say? ‘Don’t worry, your data is safe even though it’s within
easy missile range of Syria, Hezbollah, Hamas, Iran and a whole range of
terrorists.’”
From there, it was on to Check Point and then to EMC, one of the
world’s leading companies in informa on infrastructure solu ons. His
latest job change took him to Google in Mountain View, California. He
hopes to return to Israel in the next few years.

Talpiot Class 14
Ofir served in a technology unit in the IDF and later became
one of Talpiot’s wealthiest graduates. Along with fellow
classmates, he developed XIV (named a er their famous Class
14 of Talpiot), a high-end data storage system that caught the
eye of execu ves at IBM; then it caught their checkbook. IBM bought XIV
for $300 million in 2008. At the me, it was the largest acquisi on ever of
an Israeli company.

Talpiot Class 14
Barak is a rarity among secular Israelis. He grew up in
Jerusalem’s old city. His family had moved there a er the 1967
Six-Day War, for they considered it their patrio c duty to live in
areas were Jews were once forced out. Coincidentally, when
he was fi een he moved to Jerusalem’s Talpiot neighborhood, a part of
West Jerusalem that is more modern than the old city.
Instead of going for the big money a er his army service, Barak joined
Israel’s na onal police force. He sees crime and corrup on – and the lack
of respect for the country’s laws and ins tu ons – as the biggest threat the
na on faces; bigger than Israel’s external enemies, bigger than the threat
from Iran.
Barak didn’t become a beat cop or a traffic officer. His goal was to come
up with so ware that would modernize Israel’s outdated way of keeping
records. He spent five years working in the police force, upda ng systems
to make the Israeli police more efficient. And his legacy lives on. Since
leaving the force, two other Talpiot graduates signed on to add to the
effort he began.
Talpiot Class 15
Saar never wanted to use his Talpiot background to make
millions. Instead, his goal was to make a significant
contribu on, even if it meant being in a smaller environment.
He turned down a super-lucra ve job offer at Check Point,
saying the company was just too big for him. Efforts of the
Talpiot network to recruit him failed un l Saar accepted an offer at EMC in
Beer Sheva. He has been a key player in developing EMC’s Recover Point, a
program that finds lost data a er disasters, or massive hardware and
so ware malfunc ons.

Talpiot Class 15
We met Arik as the Talpiot grad who wanted to be a pilot at all
costs, and le him playing an ac ve role in the reserves,
training pilots in dogfigh ng. A er leaving the army, he started
Metacafe, an early video-sharing website. In essence, it’s a
higher quality version of YouTube, with professionally produced videos and
Internet-based television shows. He was able to secure $3 million in
financing from two of the biggest names in Silicon Valley venture capital,
Accel Partners and Benchmark. Arik sold his stake for $2.5 million.
These days, he looks out at the bright blue Mediterranean from his
eleventh-floor office in a building that houses the second company he
created, Supersonic Ads. The view is stunning, his office is casual; it’s the
office you might expect for a highly educated fighter pilot turned Internet
execu ve. There’s a large model of the anatomy of a Great White Shark, an
electric guitar, a few pairs of jeans and sneakers thrown about the room, a
heavy green coat and the na onal jersey of the US soccer team. Supersonic
Ads is a global leader in online video game adver sing, social networks and
direct response adver sing. The company also mone zes virtual currency.
It is a key component in Facebook’s ever-popular Farmville game. Any me
you want to buy something for your use in the game, you go through
Supersonic Ads to turn your real money into Farmville dollars. The
company’s ads hit 500 million social gamers throughout the world.
In his spare me – of which there isn’t much – Arik found a way to
crea vely organize several Talpiot alumni events each year. Graduates give
speeches to their fellow Talpiots about the new technologies they’ve
developed and are working on in the private sector. The grads use it as a
con nuing educa on series.

Talpiot Class 16
One of several unofficial Talpiot historians, Amir is one of
Talpiot’s most popular graduates. A er Talpiot, he tried to be a
fighter pilot, but didn’t quite make the cut. Invited to
helicopter pilot training school, he declined, saying, “No
offense to my many friends who are chopper pilots, but for me it was
fighter pilot or bust.”
A er comple ng eleven years of service in the army’s research and
development unit and serving as a Talpiot commander, he was accepted to
business school at Harvard, Wharton, MIT, Columbia and INSEAD. He chose
the Paris-based INSEAD, fearing that the American schools led down only
one path – to American corporate ins tu ons. Amir didn’t want to live
abroad; he wanted to help develop business in Israel.
Work as a consultant at McKinsey & Company, was followed by a
posi on at Israel’s biggest bank, Bank Hapoalim, as the direct assistant to
the CEO, the equivalent of a corporate chief of staff. Amir then took his
banking experience to the world of global e-commerce, founding “Global
E,” designed to make interna onal banking easier and more efficient than
ever before.

Talpiot Class 18
Argen nian-born Adam served in a technology unit of the
Israel Intelligence Corps a er gradua ng from Talpiot. A er
nine years as a so ware warrior for the Israeli army, he took
his skills into the private sector.
A er crea ng so ware designed to make personal mobile devices
compa ble with all pla orms at work, saving companies billions in
hardware costs, he started work at a new company called Screenovate.
Intel is an early financial backer; tech powerhouses Nvidia and Samsung
have also signed on to cooperate. Screenovate allows any smart phone to
turn your TV into a smart TV. Screenovate users can take whatever is on
their mobile device and put it on a big screen television with a simple click.
It’s great for work presenta ons, where you can program or save a
presenta on on your phone, then put it on a big screen for all to see in a
mee ng. The company is also marke ng products for auto companies for
dashboard displays, for gaming companies, and for home entertainment.
Kariv also used his programming skills to come up with the Public
Knowledge Workshop, a free online database that allows the Israeli public
to see exactly where the government is spending money, among other
important facts and figures about the Knesset, Israel’s parliament.

Talpiot Class 26
One of the few women to be recruited to her Talpiot class,
Marina’s army training was in the especially intense unit of the
Giva Brigade. A er her basic Talpiot training and educa on
(primarily in physics, math and computer science), she spent a
good deal of me on the Ofek satellite project.
While working in research and development, she also became a big part
of the Talpiot recrui ng system. Today she travels interna onally, speaking
to women’s groups on the importance of women in the Israeli Defense
Forces and about gender equality in the IDF.

Though not a Talpiot grad, General Ben-Israel was featured in


chapter 7 as an award-winning Talpiot role model. From high-
level posi ons in the intelligence and weapons development
units of the IAF, he went on to leadership of MAFAT and took
charge of Talpiot. Beyond his military achievements, he has
also become a highly valued asset to Israel’s business
community. He is a former member of the board of directors at Israeli
Aerospace and Industries, the country’s biggest and most prolific defense
contractor (and a leader in Israel’s unmanned drone projects). Ben Israel
also served on the advisory board at Teva, the world’s largest maker of
generic pharmaceu cals. Teva and IAI are Israel’s two biggest employers.
Ac ve in the Kadima party of Ariel Sharon and Ehud Olmert, the general
was a member of the seventeenth Knesset (Israeli parliament).

But there is yet another Talpiot graduate at this reunion whom


we have not met before. Tall and thin, Yossi Azar looks nothing
like a warrior. He looks every bit like a winner of the “Math
Olympics” (first prize in London) and the head of Tel Aviv
University’s pres gious Department of Computer Science.
Recruited to the third class of Talpiot in 1981, Yossi was not
a typical soldier – not even a typical Talpiot soldier. By nature bookish and
reserved, he says, “Talpiot changed my personality. You learn to belong to
a group, and it’s a great feeling. I became much more team oriented.”
He wasn’t quite ready for the rigors of basic training. Because he
suffered from asthma, the IDF screening doctors sent him to a less intense
training course than the one other Talpiots had to endure. With a wide
smile, Azar explains how he beat the system. “I sought out a medical
commi ee and pleaded with them to take me off the asthma list so I could
do a regular basic training. Looking back on it all thirty years later, it isn’t so
easy to understand why I did it. But I was eighteen and very mo vated to
be part of the group. The other guys thought I was crazy to ask for the
tougher training. It was very hard for me, but I made it through.”
According to the professor, in Talpiot’s eyes Israel had enough fighters
and enough academics and engineers. “What they needed was people like
us who could straddle both worlds, understand both worlds and connect
both worlds.”
He notes that the IDF was becoming increasingly compartmentalized.
Talpiot’s founders were among the first to see this trend and realized the
importance of grooming the right leadership for it. At eighteen, he didn’t
understand the need for this, but looking back on it, he says, “I was wrong.
The IDF was way ahead of me. Talpiot was able to become a new and
necessary leader in that new structure.” Yet even as he was gradua ng
from the program there were s ll some detractors high in the IDF brass
who would say the program was too expensive and wasn’t necessary.
Upon gradua on, Yossi was assigned to an intelligence unit. Using math
and computers to solve problems was his goal from the start. “I always felt
it was important work that I was doing, even though it was academic
work,” he a ests. Even twenty-five years later, he s ll wouldn’t talk about
most of the work he did for the unit.
He met his future wife in that same intelligence unit. The two later
moved to California where Yossi studied at Stanford before moving to
Redmond, Washington to work on various projects for Microso . It was
there that he met many Iranian so ware engineers. The experience shot a
chill up his spine. He saw first-hand how bright the Iranian scien sts are,
and though many were an -regime, some were not. He knew then that
superior brain power would lead to a more powerful Iran in the future.
“Not such a great thing for Israel or for the free world,” he adds
though ully.
And that is just one of the reasons Israel needs the edge provided by
Talpiot. Yossi is certain that the program is cri cally important to combat
such threats, to fuel the economy and to keep Israel ahead of the
technological game. “The program gets mo vated people and makes them
even more mo vated; and that’s great for all sectors of Israel. Just look at
the special quali es of the graduates. They succeed in the military and
then go on to excellent places in academia, industry, start-ups, big
companies, small companies – and some stay in the IDF. This is all good for
the country.”
CHAPTER 20
THE FUTURE

Twhile
he future, even the near future, in Israel is never clear. Poli cs change;
the right and center right have been dominant for more than a
decade, shi s in Israeli poli cs are inevitable. Peace talks with the
Pales nians and the greater Arab world come and go. Even with the most
savvy analysis, the impact of the Arab Spring – as well as new explosive
upheavals, including ISIS – in the Middle East is unpredictable. Israel’s
rela onships with the rest of the world are constantly swinging up and
down.
Internally, there are budget problems as well. In 2013, the IDF was
forced to lay off hundreds of army careerists. Former chief of staff Benny
Gantz, whose term came to an end in February 2015, assigned 250 people
the task of finding ways to save money, as the IDF was running a 20 billion
shekel deficit. In 2017, a foreign aid agreement with the United States
comes to an end. And that deadline is ge ng closer during an era of
changing foreign policy priori es in American poli cs. While US law says
the United States must help Israel maintain a qualita ve military edge over
its neighbors (the Naval Vessel Transfer Act HR 7177 of the 110th Congress,
and others), when it comes to spending bills in Congress, nothing is
guaranteed anymore.
The current leadership of Israel’s Ministry of Defense, Moshe Ya’alon, a
former chief of staff, clearly understands the challenges. In a speech in
2013, he emphasized the need to maintain a massive technological edge
over other na ons in the Middle East, saying priority must be given to
precision firepower, unmanned aircra and other unmanned defense
systems, intelligence capabili es and cyber-warfare.
Efficiency is a key component to keeping Israel safe – efficiency in
manpower, expenditures, advanced educa on and technological
deployments. Fortunately for Israel, efficiency isn’t an area where the
Arabs excel. Third class Talpiot graduate, MAFAT consultant and bio-science
execu ve Dror Ofer says, “The Arabs around us are declining.… We have
the upper hand in efficiency.” He explains this statement with an equa on,
as perhaps only a Talpiot-trained bio-scien st/military expert would:
“Imagine all Arab armies were very efficient, running at 80 percent of
op mal. Since the maximum efficiency is 100 percent – even if our army
reached that, we would be only 1.25 (=100/80) mes more produc ve
than they per soldier. Since their numerical advantage over us is much
greater than 1.25, they could easily beat us with an army twice as large as
ours. Now, imagine Arab armies are extremely inefficient, running at 1
percent efficiency. If our army is 10 percent efficient, it is s ll very
inefficient per soldier, but ten mes be er than the Arabs – thus our upper
hand.” He adds wryly, “If we were surrounded by 300 million Swiss, we’d
be in big trouble.”
Just a few years ago, in a bid to increase efficiency and to emphasize the
importance of military technology, the IDF changed the designa on of
soldiers using computers to bring down enemy infrastructure to “combat
soldier.” A good deal of specula on erupted in the Israeli media, and an
ar cle in the Jerusalem Post surmised that the designa on was an
admission “that Israeli cyber units are used for a ack purposes and not
only defensive purposes.” The IDF had no comment.
The glue that s cks together all of these unique resources for the IDF is
Talpiot, though the expense of the program is always under scru ny. Arik
Czerniak, (part me fighter pilot and full me high-tech entrepreneur)
discussed the rela vely small investment the IDF is making in Talpiot
cadets, in return for significant gain. “Talpiot is absolutely necessary for the
future. What is the army doing? Paying for thirty or forty kids to go through
academic training? It’s negligible. They’re building a team of people who
are really commi ed to research and development. One air force pilot
burns more money in a week than the en re Talpiot program costs for a
year. So I don’t see it as an expensive program. It’s cheap, but structured in
a very smart way, making the return on investment very high.”
On what future projects will Talpiot grads likely be u lized? They will
certainly be expected to work formally and more closely with the Combat
Intelligence Corps. It’s composed of three different ba alions including the
Shahaf (Seagull) which works with the northern command, the Nesher
(Eagle) in the very busy south with its border at Gaza, and the Nitzan
(Blossom) in the central part of the country, which includes the West Bank.
These units work with complex new technologies designed to monitor and
intercept communica ons of the enemy without pu ng soldiers in harm’s
way. Closed-circuit television cameras are designed to alert troops on the
ground when suspected terrorists, or anyone else, is nearing the border or
designated de-militarized zones. These technologies are especially valuable
to Israel because the IDF no longer needs troops in these danger zones
patrolling the border. In the past, soldiers on the border have been si ng
ducks for terrorists. Now those soldiers can respond to alerts given to them
by the Combat Intelligence Corps while wai ng in rela ve safety.
The commander of the Nesher unit told the Jerusalem Post in 2013,
“We’re known as Gaza’s big brother. We see everything. We’re seeing
Hamas ge ng stronger and preparing itself. We see them watching us.”
During the war in Gaza in the summer of 2014, known in Israel as Solid
Rock (and abroad as Opera on Protec ve Edge), Hamas u lized a new
offensive weapon: tunnels. It was a weapon the terrorists had employed
before, using their underground network in the 2006 a ack that led to the
capture of Gilad Schalit and le two Israelis dead. But in 2014, Hamas
began using these tunnels as a larger part of their strategy. Many Israeli
military analysts commented Israel had the advantage in the air, at sea and
on the ground, but not underground. Another joked, “Israel should hire
Hamas to build the Tel Aviv subway.”
But those tunnels were certainly no joking ma er. Some stretched from
Gaza deep into Israeli territory, surfacing near kibbutzim and other civilian
infrastructure. Israeli intelligence discovered that the tunnels were meant
to be used during the fall of 2014 in “a mega terror a ack” around the
Jewish holidays. Inside one captured tunnel, Israeli forces discovered
tranquilizers and handcuffs that would have been used to bring Israelis
back into Gaza as hostages.
Despite Israel’s knowledge of the tunnels, not enough was done to
defend against them in 2014. Hamas used them to successfully carry out
several a acks in Gaza and inside Israel proper, resul ng in the deaths of
several Israeli soldiers.
As the war was raging fi y-three miles to the southwest of Jerusalem,
the Knesset’s Science and Technology Commi ee finally began looking for
ways to combat the tunnel threat. The head of the commi ee, Moshe
Gafni of the Degel HaTorah party, called for the commi ee to address the
tunnel threat immediately a er the conclusion of the 2014 war, saying they
expect to hear tes mony and ideas from experts with experience in
geology, mining and other similar fields.
Talpiot also began working on solu ons. The tunnel threat is likely to be
the focus of future second-year student projects that will be presented to
senior army officers. While it’s impossible to say what a future solu on will
look like, everyone agrees that for life to con nue with some semblance of
safety and security, especially in Israel’s southern towns, something must
be done. Ideas have already surfaced regarding underground warning
sensors, but nothing concrete has been decided.
Avi Isaacharoff, a reporter and analyst for The Times of Israel and
Walla.co.il (one of Israel’s top Internet portals) and author of the book 34
Days about the Second Lebanon War, surmised in a Times of Israel ar cle
(August 2014), “Hezbollah was the first to use defensive tunnels to a ack
IDF soldiers and to con nue firing rockets. One can assume that, in the
eight years since the Second Lebanon War, Hezbollah has accelerated its
digging project on two levels: defensive tunnels within Lebanon and a ack
tunnels into Israel. Eight years of work can mean that, in the next war, we
will find Hezbollah fighters emerging from tunnels deep inside Israel, not
necessarily near the border. Yes, the ground there is harder to excavate
compared to that of Gaza, but it can be assumed that, as always, what
Hamas does well, Hezbollah does be er.”
It’s clear that without a solu on from Talpiot or somewhere else, the
frightening prospect of more tunnel a acks in the south and or the north
could end in tragedy for Israel. In many regards, a technical solu on is only
the first step for Israel in dealing with the tunnel threat. In a news
conference (August 6, 2014), a er a three-day cease-fire with Hamas went
into effect, Prime Minister Netanyahu dedicated much of his address to the
na on and the world media to that tunnel threat: “Israel is working to
create technological means to locate new tunnels that will reach into our
territory.”
Talpiot-trained soldiers also used their technology know-how to
construct dozens of remote controlled guns sta oned on the Israel-Gaza
border that are operated from a command post several miles away, o en
by specially trained female troops. These guns also help cut back on the
number of ground troops Israel needs to make sure that terrorists don’t
cross the line.
While border ba les are certainly a priority, so is longer range figh ng.
Talpiot graduates are involved in preparing the Israeli Air Force for
American made F35s, which will replace the older F16s. One major tool
F35 pilots will possess is the helmet display system designed by the Israeli
firm, Elbit Systems. Elbit and its staff of Talpiot engineers are working with
the US defense firm Rockwell Collins on the project. The fighter pilot
helmet of the future gives a flier pictures from on-board cameras giving
him a be er look at targets, objec ves and obstacles, as well as the
mechanics and performance of the airplane.
Famous Talpiot graduate and head of MAFAT Ophir Shoham is inves ng
more and more in robo cs. He doesn’t believe robots will ever replace
soldiers in the Israel Defense Forces. But he is convinced that by 2020
robots will have more combat roles and will cut down on Israeli casual es
by going first into hos le territory. He told Haaretz correspondent Amos
Harel in 2012, “More robots will not replace warriors.… But unmanned
vehicles on the ground will go to high-risk targets, you can send them from
afar into enemy territory, a kind of front guard, vehicles that watch a
situa on and shoot. This will happen in the foreseeable future.” Shoham
envisions a greater presence of robo cs on the ground, unmanned ground
vehicles (UGVs) almost equal in a sense to Israel’s world-class fleet of
unmanned aerial vehicles.
The head of MAFAT’s robo cs division is Lieutenant Colonel Gabi
Dobresco. He says many UGVs are already in opera on; several are helping
with surveillance along the Gaza border, others are aiding troops near Arab
areas of the West Bank. He believes UGVs will be used more in the future
to find and detonate roadside bombs and landmines. They’ll also be used
in urban areas to draw fire, helping the IDF locate the enemy. In a release
issued by the Ministry of Defense, Lieutenant Colonel Dobresco said,
“Robots some mes go in front of the forces to open challenging roads such
as narrow alleys and assist logis cally. A robot can help lighten a soldier’s
burden, so that if the soldier is confronted with a ba le, he or she can
respond appropriately.” He also said that in the future, “the UGV will be
equipped with obstacle detec on sensors, cameras and other tools, and it
will be able to iden fy the barriers by itself and circumvent them.” Talpiots
are also at the forefront of technology when it comes to the development
of both air-and ground-based remote-controlled figh ng vehicles.
Talpiot graduates will also con nue to have a major impact beyond the
IDF, especially when it comes to taking the Israeli economy to new heights.
During a trip to Jerusalem in the summer of 2013, Cisco’s CEO John
Chambers said that Israel will be the world’s first digital na on. Israeli
innovators and entrepreneurs, including many Talpiot graduates who are
either building the networks or helping to fund them, are connec ng more
and more sectors of the economy to fiber op cs. It will have an impact on
the healthcare system, the economy, and how people at home
communicate, learn and do business.
One area where Talpiot students have not yet had an impact is in the
world of poli cs. To date, not a single Talpiot graduate has made waves in
Israel’s poli cal system, but that may one day change. Every Talpiot cares
deeply for the country and it’s likely that just as they’ve changed the
military establishment, a Talpiot graduate may very well change the
poli cal landscape as well.
Though Iran con nually threatens “Israel will be wiped off the map,” few
of the Talpiot graduates interviewed for this book believe Iran is Israel’s
biggest problem. Colonel Avi Poleg, who turned his army and Talpiot
experience into a career in global private educa on, laughed when asked
about the rising power of Iran. He says, “Iran is not a military problem for
Israel right now. I’m not worried about Iran’s bomb or the Pales nians. The
most fearful, problema c thing to me, is the process going on in our
society. We stood against very severe challenges in our history because we
were strong from the inside. If you see cracks in the internal workings of a
society, this is most dangerous.”
One of the “cracks” he is alluding to is the ri between the growing
Haredi popula on and general society. Popular resentment boils over into
bi er acrimony in the press because Haredim have been exempt from
military service since 1948 (when their numbers were compara vely few).
Hot debate over their proposed induc on into the army only widens the
cultural divide, weakening the unity of the na on. Talpiots are also more
worried about an unbalanced economy, as well as internal Israeli crime and
corrup on than they are about an Iranian nuclear bomb.
But societal concerns don’t grab headlines the way military threats do.
Barak Ben-Eliezer, told The Marker, “Educa on and social issues are
perhaps less photogenic, but just as acute of a problem as security. They
are internal bleeding issues, rather than external bleeding, and therefore it
is more difficult to catch them and expose them.”
His allusion to educa on as another of the “cracks” threatening Israel is
surprising. But Israeli educa on isn’t what it once was. In a recent study of
teenagers in sixty-five “developed” countries, Israel ranked 41st, alongside
Croa a and Greece in mathema cs. In science, Israel also ranked just 41st.
If there are two areas where Israel simply can’t afford to fall behind, it’s
mathema cs and science. Both are cri cal to a country that prides itself on
innova on – and it needs innova on to survive.
Perhaps a Talpiot-like approach is needed to fix the educa on system in
Israel. It’s a method already championed by Colonel Poleg, who uses
Talpiot as a model when he consults in schools both within and outside of
Israel.
David Kutasov, a mild-mannered, so -spoken string theory physicist at
the University of Chicago, describes his job by asking, “Did you ever watch
the TV show The Big Bang Theory? I have the same job as Sheldon Cooper,
the weird skinny guy. My job is to find out how the universe came into
being.”
Kutasov believes Talpiot will con nue to lead in research. And he knows
why. “Research through Talpiot is o en about originality. I would not have
been as good of a researcher or a physicist without Talpiot. It empowers
you.
“A lot of kids I see these days – even at top American universi es – are
too conven onal and not original. At Talpiot, they beat it out of you and
push you toward originality. Now let’s look at the American system. My
daughter was admi ed to MIT for engineering. But of her class, only she
wants to be an engineer. The rest want to get their MBA, but they stayed
with the herd and applied to MIT. Another example: in Manha an you
need the right preschool to get to Dalton, to get to Harvard, to get to the
right law school. The system breeds unoriginal professionals, and it only
gets you so far.
“It seems the most important tech leaders in the US didn’t finish college
at all. Steve Jobs dropped out. Bill Gates dropped out. Look at all these
MBAs be ng on credit default swamps. Didn’t anyone ask ‘Is this a good
idea?’ The system breeds followers and not leaders.
“On the other hand, Talpiot consistently breeds leaders. Talpiot
emphasizes originality. They bring in people to tell you about what’s going
on in some branch of the army, then ask you how you would do it
differently. They keep challenging you all the me. It’s in the genes of the
program.”
From Talpiot’s incep on, those “genes” have propelled the program and
its graduates to unexpected and unprecedented breakthroughs. One day,
these young men and women will undoubtedly start to influence Israel’s
government and policies, and perhaps even to enhance prospects for
peace. As Talpiot alumni expand their influence in all sectors – from
educa on to defense, to the halls of power in Jerusalem – there is hope
that Israel will be safe and secure for many genera ons to come.
Thank You for Downloading this Gefen Publishing House Ebook.

Sign up for our free weekly newsle er and get special offers on
exci ng new tles about Israel, Judaism and the Jewish People.

To sign up click here!


APPENDIX
TIMELINE

DATE HISTORICAL EVENT TALPIOT EVENT TALPIOT


GRADS’
ACHIEVEMENTS
1918/1919 End of World War
I. End of O oman
rule in Pales ne.
League of Na ons
grants mandate to
Great Britain to
rule Pales ne.

1924 Felix Dothan born in Yugoslavia.

1927 Shaul Yatziv born in Bri sh Mandate


Pales ne.

November United Na ons


29, agrees to par on
1947 Pales ne between
Arabs and Jews.

May 14, State of Israel


1948 established. US
recognizes Israel.

1948–1949 Israel is a acked


by surrounding
Arab na ons. War
of Independence
ends in armis ce
agreements with
Egypt, Jordan,
Syria and
Lebanon.
Jerusalem is
divided between
Israel and Jordan.

October Sinai Campaign:


1956 Israel takes Sinai
with Bri sh and
French support.
Sinai returned to
Egypt under
pressure from the
UN, US and USSR.

1950s Ongoing Arab


/1960s guerrilla a acks;
IDF reprisal
opera ons.

June 1967 Six-Day War. Egypt


blocks the Straits
of Tiran. Israel
captures Sinai
from Egypt; the
West Bank,
including
Jerusalem, from
Jordan; and the
Golan Heights
from Syria.

November United Na ons


1967 Resolu on 242
adopted as a
framework for
peace.

1967–1970 Egypt/Israel War


of A ri on.

October Yom Kippur War.


1973 Israel is a acked
by Egypt and
Syria; sustains
ini al losses but
ul mately defeats
Egypt and Syria.
Cease-fire
declared.

April 1974 Interim report of


the Agranat
Commission
inves ga ng the
Yom Kippur War
fiasco cites a lack
of military and
intelligence
preparedness.
Resigna ons of
several top
military leaders, as
well as Prime
Minister Golda
Meir.

November Professors Shaul Yatziv and Felix


1974 Dothan formulate a document tled
“A Proposal for Establishing an
Ins tute for the Development of
New Weapons.” Today it is simply
called “The Ini a on Document” of
the Talpiot program. Dothan and
Yatziv’s original proposal was to
grant degrees in physics and math to
gi ed recruits within a span of
twelve months, and a erwards
direct them to crea ng solu ons for
the army.

1974 Intel opens Israel


office, its first
development
office outside of
the United States.

July 1975 A mee ng at the Ministry of Defense


considers the crea on of the Talpiot
program. At the end of the mee ng,
the par cipants agree it is a good
idea, but several ques ons remain
and there is no formal approval.

1971–1982 PLO is expelled


from Jordan in
1971 and sets up
base in southern
Lebanon. It stages
a acks on Galilee
residents and is
countered by IDF
retalia ons.

March Opera on Litani.


1978 Israel invades
Lebanon on a
large scale for the
first me. The PLO
withdraws from
southern Lebanon
and sets up base
north of the Litani
River. A UN buffer
zone is established
on the Israel-
Lebanon border.

September Camp David


1978 Accords are signed
between Egypt
and Israel.
Egyp an president
Sadat and Israeli
prime minister
Begin share the
1978 Nobel Peace
Prize.

1978 Under Prime


Minister Begin,
General Rafael
Eitan becomes
Israel’s 11th chief
of staff, launching
an era of
innova on.

March Israel-Egypt Peace General Eitan officially approves


1979 Treaty signed. Talpiot and gives ini al instruc ons
for implemen ng the program.

spring Recrui ng for Talpiot begins and


1979 plans for the program are
developed. The first Talpiot class
reports for duty in the summer of
1979.

1981 Talpiot’s base is moved from the


Palmachim military base to Hebrew
University in Jerusalem.

1978–1982 PLO con nues its


campaign against
Israel. Israel
invades Lebanon
again in 1982 and
forcibly expels the
PLO. Israel
withdraws to a
slim borderland
buffer zone, held
with the aid of the
South Lebanon
Army (SLA).

1981 Ariel Sharon


named Minister of
Defense.

1982 Israeli Navy is recep ve to Talpiot


cadets; looks forward to Talpiot help
in modernizing. First Talpiot soldiers
graduate from the three-year
academic program.

1982–2000 Ongoing war in Talpiot cadets are sent to Lebanon


southern Lebanon for officer and combat training.
between the IDF,
Hezbollah and
other mili as and
guerrillas.

1983 Computer science is added to


Talpiot’s curriculum.

1983 Israel Space


Agency founded
by Yuval Ne’eman,
who headed the
Ministry of
Science and was
instrumental in
ge ng Talpiot
approved by the
Ministry of
Defense.

1984 Talpiot begins allowing women into


the program.

1985 Hezbollah, a
Lebanese Shia
radical movement
sponsored by Iran,
calls for armed
struggle to end
the Israeli
occupa on of
Lebanese territory.

mid-1980s Talpiot Professor Ariel Lorber devises


concept and original proposal for
Trophy ac ve protec on system for
armored figh ng vehicles.

1985 Talpiot graduates become instructors


in the Talpiot program, commanding
at class level, for the first me.
Talpiot graduate, Opher Yaron,
becomes the first graduate to lead
the program. Women are more
aggressively recruited to Talpiot.

1987–1993 First In fada –


Pales nian
uprising against
Israel in the West
Bank and Gaza.

1988 Israel launches its


first Ofek satellite.

1991 Missiles are Classes in military studies become


launched at Israel mandatory for Talpiot students.
from Iraq during Courses include instruc on in the
the “Scud War.” Arab-Israeli conflict, the modern
ba lefield and na onal security.

1993/1995 Oslo Accords


reached between
Israel and the PLO.

mid-1990s Rockets are fired Talpiot students suggest ini al


at civilian centers concept for Iron Dome defense.
in the south of
Israel.

1993 Check Point


So ware
Technologies
co-founded by
Talpiot grad
Marius Nacht.
Compugen
founded by
Talpiots Eli
Mintz,
Simchon
Faigler and
Amir Natan.

October Israel-Jordan
1994 Peace Treaty
signed.

1994 Era of suicide


bombings begins.

November Assassina on of
1995 Prime Minister
Yitzhak Rabin.

1996 Check Point


So ware
Technologies
lists on the
Nasdaq.

1999 Twen eth anniversary of Talpiot


celebrated.

2000 Talpiot program enlarged to Compugen


accommodate almost twice as many lists on the
cadets. Nasdaq.

2001 Passave
founded by
Talpiot grad
Ariel Maislos.

2003 Talpiot tes ng and recrui ng is X Technologies


redesigned under Commander Amir purchased by
Schlachat. Guidant.
Talpiot grad
Guy Shinar is
closely
associated
with the
company and
with this first
very large
Israeli biotech
deal.

2000–2005 Second In fada –


Pales nian
uprising against
Israel in the West
Bank and Gaza.

July– Second Lebanon


August War against
2006 Hezbollah

2007 XIV, founded


by members
of Talpiot
Class 14
including
Barak Ben-
Eliezer and
Ofir Zohar, is
sold to IBM.

2008 Talpiot’s thir eth class begins.

December First Gaza War


2008– (“Opera on Cast
January Lead”) against
2009 Hamas.

2011 Israel’s Cyber War Waves Audio


Bureau is founded Ltd., founded
as an advisory by Talpiot grad
mechanism to the Meir
prime minister Sha’ashua,
and to develop wins Technical
cyber defense. It is Grammy
headed by Talpiot Award.
graduate Eviatar
Matania.

January Anobit, co-


2012 developed by
Talpiot grad
Ariel Maislos,
sold to Apple.

November Second Gaza War


2012 (“Opera on Pillar
of Defense”).

July– Third Gaza War


August (“Opera on
2014 Protec ve Edge”).
A SALUTE FROM THE AUTHOR

Any me a book or detailed ar cle is wri en about a sensi ve military


program, the writer owes a debt of gra tude to a lot of people who could
not be named, but helped immensely with the research for the project.
Israel has strict security rules in place to protect the iden es of its pilots,
espionage agents, military analysts, defense execu ves, officers, fighters
and scien sts. The IDF’s Talpiot program is an all-star list made up of all of
those components so necessary for an effec ve and formidable figh ng
force.
I’d like to thank the men and women from those ranks who helped me
with informa on and research, and who simply helped me understand
how the IDF machine works. While many could not be named for security
reasons, and others did not have proper permission to speak to a
journalist, their contribu ons were vital to this book.
This list includes men and women formerly of the Mossad who were so
helpful with their me; execu ves from Israel’s military contractors, large
and small; former members of the IDF’s spokesperson’s office who were
generous with their me but s ll needed to obey the rules of secrecy on
certain subjects; current field commanders; air force pilots; former naval
officers; several ac ve Talpiot soldiers; and several Talpiot graduates now
serving in various branches of the IDF and the IDF reserves.
They have served and con nue to serve their country so bravely –
commanding elite units in the field, always preparing for the next war and
always trying to stay ten steps ahead of the enemies surrounding Israel.
Fortunately, there are many who graciously allowed me to use their
names. Many thanks to the following Talpiot graduates:
Gilad Almogy
Mor Amitai
Matan Arazi
Yossi Azar
Uri Barkai
Oded Bar Lev
Ziv Belsky
Ron Berman
Saar Cohen
Arik Czerniak
Tal Dekel
Zvika Diament
Rotem Eldar
Barak Ben-Eliezer
Elad Ferber
Gideon Fos ck
Marina Gandlin
Raanan Gefen
Kobi Kaminitz
Adam Kariv
Opher Kinrot
Giora Kornblau
Ophir Kra-Oz
David Kutasov
Gilad Lederer
Ron Milo
Eli Mintz
Marius Nacht
Dror Ofer
Amir Peleg
Barak Peleg
Avi Poleg
Boaz Rippin
Amir Schlachet
Haggai Scolnicov
Guy Shinar
Tal Slobodkin
Dr. Aviv Tu nauer, MD
Opher Yaron
Guy Levy-Yurista
Ofir Zohar
Special thanks: Amnon Govrin, brother of Talpiot graduate
Oded Govrin (deceased)

Early Administrators/Organizers
Yoav Dothan, son of Felix Dothan, Talpiot’s founder
General Uzi Eilam, former head of MAFAT
General Yitzhak Ben-Israel, former head of MAFAT
Colonel Benji Machnes, early founder, from the air force
Hanoch Tzadik, early administrator of the program, serving as a
psychologist

Early Observers of the Program/Former Soldiers


Odelliah Cohen
Meidad Muskal

Government Employees, Ministry of Foreign Affairs


Oren Anolik
Benjamin Krasna

Army Spokesperson’s Office


Captain Eytan Buchman
Captain Keren Hajioff
Colonel Avital Leibovich
Libby Weiss
Lt. Col. Limor Gross-Weisbuch

Defense Contractors
David Ishai, Rafael
Noga Nadler, IAI

Industry Contacts
Mar n Gerstel, Venture Capitalist
Yossi Gross, Rainbow Medical
Sharona Justman, STEP Strategy Advisors

Professors/Historians/Journalists
David Horovitz
Arieh O’Sullivan
Abraham Rabinovich
Amir Rapaport
Professor Jonathan Rynhold
Colonel Shaul Shay
Ronen Bergman
Ron Schleifer

Pop Culture Observer


Asaf Harel actor/producer/writer

The Staff at the IDF Archives Israel Consulate, New York


Shahar Azani
Keren Gelfand
ACKNOWLEDGMENTS

There is no way this book would have been possible without Charlo e
Friedland and Yitzchak Friedland. Their pa ence in teaching me how to
structure the book and their help in reworking it were simply invaluable to
me. I will never be able to properly thank them. You are both wonderful
editors and wonderful people. Thank you.
I want to thank Ilan Greenfield, Lynn Douek, Kezia Raffel Pride and
Esther Schwartz-Ivgy of Gefen Publishing House for believing in me and in
the book and of course for publishing it. I think Ilan really understood my
vision for the book and my passion for the story from the first e-mail pitch I
sent to him, many years ago.
Yehudit Singer, in Jerusalem, was a huge help connec ng me with
important people in Israel, searching for and finding impossible-to-find
documents and transla ng them. Without her, I’d s ll be si ng on a cold
library floor in Jerusalem with my English-Hebrew dic onary.
I’d like to thank many of the people at the IDF spokesperson’s office,
including Colonel Avital Leibovich, Colonel Limor Gross-Weisbuch, Eytan
Buckman, Keren Hajioff and Libby Weiss. At the Ministry of Defense, I will
be forever in the debt of several people who not only gave me advice, but
served as valuable fact checkers…one even taught me how to walk
between the raindrops; thank you.
Ronen Bergman, one of Israel’s brightest military journalists, was
extremely generous with advice and ideas along the way.
Hadassah Medical Center’s chief public rela ons execu ve, Barbara
Sofer, was always available for advice, counsel, friendship – and she could
always be counted on to make me laugh.
Jeffrey Gewirtz and Stacy Gewirtz were terrific assistants, traveling with
me to and from Israel. My father, Michael Gewirtz, was able to provide me
with more research than I’ll ever be able to properly organize and store; he
was reless in his efforts to suggest material he found online and in
newspapers across the globe.
I want to thank Robert DeFelice for his help in graphically explaining my
ideas for the cover so that Gefen’s ar sts were able to produce the vision
that I was unable to put into words.
Ron Schleifer of Israel’s Ariel University was also very helpful with
sugges ons on how to turn my idea into a book, where to start and how to
proceed throughout the way.
I’ve been extremely fortunate to work with some of the best people in
the news business. Mark Hoffman has turned CNBC into the top brand in
business news. He’s made CNBC a place where all of his employees can
honestly say it is a place where they’re proud to work.
I especially want to thank Nik Deogun for his enthusiasm as I con nued
to research and write. Nik is one of the best and wisest newsmen I’ve ever
met. His breadth of knowledge is never-ending; not just when it comes to
content, but when it comes to management as well.
Jeremy Pink was excep onal for his ini al encouragement at the
beginning of this project. I want to thank David Friend and Joel Franklin for
bringing me to CNBC.
This book would not have been possible without Jonathan Wald, who
sent me on my first assignment to Israel as a producer and then kept
sending me to cover war, peace and Warren Buffe ’s historic trip to the
country in the fall of 2006. Those assignments gave me confidence not only
to write this book, but to grow personally and professionally. Working
interna onally – including in the Middle East during mes of war, and
during rio ng in Greece – with Carl Quintanilla taught me an enormous
amount about truly working under pressure. The me I spent with Carl in
the field were some of my proudest days as a journalist. I was also
fortunate to work with many of the people in NBC’s former Tel Aviv
bureau, including Gila Grossman, Paul Goldman, Dave Copeland and
Mar n Fletcher, one of the most accomplished journalists to ever cover
Israel. Traveling and working with CNBC’s interna onal correspondents
Michelle Caruso-Cabrera and Senior Correspondent Sco Cohn were also
experiences that had a big impact on me.
Without his knowledge, I used Ben Sherwood’s book The Survivor’s Club
as a model for parts of this manuscript. Ben is a great story teller. I learned
a lot from the short me I was able to spend with him and from his
remarkable ability to tell a story and to let subjects speak for themselves.
Jim Rivas at Check Point So ware Technologies was instrumental in
helping secure key interviews and material.
David Raab, author of Terror in Black September, was excep onally
generous with his me and advice as I began to write.
Thank you all very much.

You might also like