GitHub - anthropics/prompt-eng-interactive-tutorial: Anthropic's Interactive Prompt Engineering TutorialYou signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session. Dismiss alert
AIエージェントを脱線ã•ã›ãªã„プãƒãƒ³ãƒ—トè¨è¨ˆã®å®Ÿè·µã‚¬ã‚¤ãƒ‰
ã¯ã˜ã‚ã« Devinã‚„GitHub Copilot Agentã®ã‚ˆã†ãªAIエージェントã¯ã€è¨è¨ˆãƒ»å®Ÿè£…ã®ä¸€éƒ¨ã‚’自動化ã—開発速度を大ããå‘上ã•ã›ã¾ã™ã€‚ã—ã‹ã—ã€æ˜Žç¢ºãªã‚¬ãƒ¼ãƒ‰ãƒ¬ãƒ¼ãƒ«ãªã—ã«æŒ‡ç¤ºã‚’出ã™ã¨ã€ãƒªãƒã‚¸ãƒˆãƒªã®å‰æを誤解ã—ãŸã‚Šã€æƒ³å®šå¤–ã®å®Ÿè£…ã«è¸ã¿è¾¼ã‚“ã ã‚Šã—ã¦ç”Ÿç”£æ€§ãŒä¸‹ãŒã‚‹ã“ã¨ã‚‚å°‘ãªãã‚ã‚Šã¾ã›ã‚“。本記事ã§ã¯ã€AIエージェントãŒè¨è¨ˆã‚„開発タスクを脱線ã—ãªã„よã†ã«ã™ã‚‹ãŸã‚ã®ãƒ—ãƒãƒ³ãƒ—トã®ã‚³ãƒ„ã‚’ã€READMEã‚„docsã®æ•´å‚™ã€ã‚¿ã‚¹ã‚¯ã®ã‚¹ãƒ©ã‚¤ã‚·ãƒ³ã‚°ã€issueã®æ›¸ãæ–¹ã¨ã„ã£ãŸè¦³ç‚¹ã§å…·ä½“çš„ã«ç´¹ä»‹ã—ã¾ã™ã€‚ ガードレールã¨ã—ã¦ã®README / docsæ•´å‚™ã¨æ´»ç”¨ãƒ—ãƒãƒ³ãƒ—ト READMEを「行動ルール集ã€ã«ã™ã‚‹ プãƒã‚¸ã‚§ã‚¯ãƒˆã®README.mdã¯ã€ã‚¨ãƒ¼ã‚¸ã‚§ãƒ³ãƒˆãŒæœ€åˆã«èªã‚€å‰æ資料ã§ã™ã€‚以下ã®è¦³ç‚¹ã‚’明文化ã—ã¦ãŠãã¨ã€ãƒ—ãƒãƒ³ãƒ—トã‹ã‚‰å‚ç…§ã•ã›ã‚„ã™ããªã‚Šã¾ã™ã€‚ リãƒã‚¸ãƒˆãƒªæ§‹æˆï¼ˆãƒ‰ã‚ュメント・アセット・コードã®å ´æ‰€ï¼‰ コー
LLMを本番å“質ã«è‚²ã¦ã‚‹ PromptOps：â€100回ã®è©¦è¡ŒéŒ¯èª¤â€ã‚’支ãˆãŸä»•çµ„ã¿ã¨æ–‡åŒ–
0. ã¯ã˜ã‚ã« ELYZAã¨æ ªå¼ä¼šç¤¾ãƒžã‚¤ãƒŠãƒ“ãŒå…±åŒé–‹ç™ºã—ãŸã€Œãƒžã‚¤ãƒŠãƒ“AI Pencilã€ãŒå…¬é–‹ã•ã‚Œã¾ã—ãŸã€‚本記事ã§ã¯ã“ã®é–‹ç™ºéŽç¨‹ã«ãŠã„ã¦é‡è¦ãªå½¹å‰²ã‚’æžœãŸã—ãŸã€ŒPrompt Engineeringã€ã¨ã€ãれを支ãˆã‚‹ä½“系的ãªé‹ç”¨åŸºç›¤ã€ŒPromptOpsã€ã«ã¤ã„ã¦è§£èª¬ã—ã¾ã™ã€‚ 「マイナビAI Pencilã€ã«ã¤ã„ã¦ã¯ã€å½“社リリースãªã‚‰ã³ã«ãƒžã‚¤ãƒŠãƒ“リリースをã”覧ãã ã•ã„。 æ ªå¼ä¼šç¤¾ELYZA プレスリリース：https://prtimes.jp/main/html/rd/p/000000062.000047565.html æ ªå¼ä¼šç¤¾ãƒžã‚¤ãƒŠãƒ“様 プレスリリース：https://www.mynavi.jp/news/2025/07/post_49493.html ãªãŠã€æœ¬ãƒ–ãƒã‚°ã«é–¢ã™ã‚‹ç ”究開発ã¯ã€ãƒ—ãƒã‚¸ã‚§ã‚¯ãƒˆã«é–¢ã‚ã£ãŸ ML（機械å¦ç¿’）エンジニアã®æ›½æˆ‘部 (@sog4be)ã€å ¤ (@ozro_223)
AIãŒæŒ‡ç¤ºã«å¾“ã£ã¦ãã‚Œãªã„ï¼AIã¨å¦å®šè¡¨ç¾ã®ä¸æ€è°ãªé–¢ä¿‚
「想åƒã™ã‚‹ãªã€ã®é€†èª¬ 「ピンクã®è±¡ã‚’想åƒã™ã‚‹ãªã€ ã“ã®ä¸€æ–‡ã‚’èªã‚“ã 瞬間ã€ã‚ãªãŸã®è„³å†…ã«ã¯ã™ã§ã«ãƒ”ンクã®è±¡ãŒè¸Šã£ã¦ã„ã¾ã›ã‚“ã‹ï¼Ÿ ã“ã®ç¾è±¡ã¯ã€Œçš®è‚‰éŽç¨‹ç†è«–ã€ã¨å‘¼ã°ã‚Œã‚‹ã‚‚ã®ã§ã€ã€Œä½•ã‹ã‚’考ãˆãªã„よã†ã«ã—よã†ã€ã¨ã™ã‚Œã°ã™ã‚‹ã»ã©ã€ã‹ãˆã£ã¦ãã‚Œã«ã¤ã„ã¦è€ƒãˆã¦ã—ã¾ã†ã¨ã„ã†äººé–“ã®å¿ƒç†çš„ãªç‰¹æ€§ã‚’指ã—ã¾ã™ã€‚心ç†å¦è€…ダニエル・ウェグナーãŒç ”究ã—ãŸã“ã®ç¾è±¡ã¯ã€ç§ãŸã¡ã®æ—¥å¸¸ã§ã‚‚よãèµ·ã“ã‚Šã¾ã™ã€‚ é©šãã¹ãã“ã¨ã«ã€ChatGPTã‚„BERTãªã©ã®æœ€æ–°AI（大è¦æ¨¡è¨€èªžãƒ¢ãƒ‡ãƒ«ã€ç•¥ã—ã¦LLM）もã€ã“ã®äººé–“ã¨åŒã˜å¼±ç‚¹ã‚’æŒã£ã¦ã„るよã†ãªã®ã§ã™ã€‚ AIã®å†…部ã§ä½•ãŒèµ·ãã¦ã„ã‚‹ã®ã‹ AIモデルã¯è¨€è‘‰ã‚’「数値ã®ç¾…列（ベクトル）ã€ã¨ã—ã¦ç†è§£ã—ã¾ã™ã€‚興味深ã„ã“ã¨ã«ã€ã€Œå¥½ãã€ã¨ã€Œå¥½ãã§ã¯ãªã„ã€ã¨ã„ã†æ£å対ã®æ„味をæŒã¤æ–‡ã§ã‚‚ã€AIã®å†…部ã§ã¯éžå¸¸ã«ä¼¼ãŸæ•°å€¤ãƒ‘ターンã¨ã—ã¦è¡¨ç¾ã•ã‚Œã¦ã—ã¾ã†ã“ã¨ãŒã‚ã‚Šã¾ã™ã€‚ 例ãˆã°ï¼š ã“れらã¯ç´„92%ã‚‚ä¼¼ã¦ã„ã‚‹ã¨ã•ã‚Œã¦ã„ã¾
Customize AI responses in VS Code
Chat in Visual Studio Code can give you responses and generate code that matches your coding practices and project requirements, if you give it the right context. Instead of repeatedly adding this information in every chat prompt, you can store this context in files and automatically include it in every chat request. In this article, you learn how to use custom instructions and prompt files to cus
AWSè¨è¨ˆãƒ—ãƒãƒ³ãƒ—ト
シンプルã‹ã¤ç¶²ç¾…çš„ãªAWSè¨è¨ˆã‚’生æˆã™ã‚‹AIプãƒãƒ³ãƒ—トã®æ ¸å¿ƒã¯ï¼š æ§‹é€ åŒ–ã•ã‚ŒãŸå‡ºåŠ›ãƒ•ã‚©ãƒ¼ãƒžãƒƒãƒˆï¼šè¨è¨ˆæ›¸ã®ç« ç«‹ã¦ã¨å„セクションã®èª¬æ˜Žå†…容を明確ã«æŒ‡å®š 具体的ãªãƒ‘ラメータè¦æ±‚：抽象的ãªèª¬æ˜Žã§ã¯ãªãã€å®Ÿè£…ã«ä½¿ãˆã‚‹å…·ä½“çš„ãªè¨å®šå€¤ã‚’求ã‚ã‚‹ é¸å®šç†ç”±ã®æ˜Žç¢ºåŒ–：「ãªãœãã®é¸æŠžã‚’ã—ãŸã®ã‹ã€ã®èª¬æ˜Žã‚’求ã‚ã‚‹ 代替案ã¨ã®æ¯”較：検討ã—ãŸä»£æ›¿ã‚ªãƒ—ションã¨ã®æ¯”較をå«ã‚ã‚‹ Well-Architectedã®åŽŸå‰‡é©ç”¨ï¼šAWSã®ãƒ™ã‚¹ãƒˆãƒ—ラクティスã«åŸºã¥ãè¨è¨ˆã‚’促㙠ã“ã®ã‚¢ãƒ—ãƒãƒ¼ãƒã‚’活用ã™ã‚Œã°ã€AIã®åŠ›ã‚’借りã¤ã¤ã‚‚ã€å®Ÿè£…ã«ç›´çµã™ã‚‹é«˜å“質ãªAWSè¨è¨ˆæ›¸ã‚’効率的ã«ä½œæˆã§ãã¾ã™ã€‚何よりã€è¨è¨ˆè€…ã®æ™‚間を節約ã—ãªãŒã‚‰ã‚‚ã€ãã®å°‚門知è˜ã¨åˆ¤æ–を最大é™ã«æ´»ã‹ã›ã‚‹ã¨ã“ã‚ã«å¤§ããªä¾¡å€¤ãŒã‚ã‚Šã¾ã™ã€‚ ※以下を全é‡ä½¿ç”¨ã™ã‚‹ã¨é‡ãŒå¤šã„ã®ã§ã€è©²å½“箇所ã®ã¿ã®æŠœç²‹ã‚’推奨 以下ã®è¦ä»¶ã«åŸºã¥ã„ã¦ã€è©³ç´°ã‹ã¤å®Ÿè£…å¯èƒ½ãªAWSアーã‚テクãƒãƒ£ã‚’è¨è¨ˆã—ã¦ãã ã•ã„。å„セクシ
æ¥å‹™æ”¹å–„ã«å½¹ç«‹ã¤ChatGPTã®ãƒ—ãƒãƒ³ãƒ—ト文例集｜ä¸å°è¦æ¨¡æ³•äººã€ä¸å …・ä¸å°ä¼æ¥ã®ãŠå®¢ã•ã¾ ｜ ソフトãƒãƒ³ã‚¯
\r\n<li><span class=\"biz-smb-fs-m1\"><b><span class=\"biz-smb-txt-link\"><a href=\"#prompt\">プãƒãƒ³ãƒ—ト一覧</a></span></b></span></li>\r\n<li><span class=\"biz-smb-fs-m1\"><b><span class=\"biz-smb-txt-link\"><a href=\"#move\">ChatGPT活用動画</a></span></b></span></li>\r\n<li><span class=\"biz-smb-fs-m1\"><b><span class=\"biz-smb-txt-link\"><a href=\"#blog\">関連記事</a></span></b></span></li>\r\n<li><span class
医療分野ã«ãŠã‘ã‚‹LLMã®ç¾çŠ¶ã¨å¿œç”¨å¯èƒ½æ€§ã«ã¤ã„ã¦
GeoCLIP: Clip-Inspired Alignment between Locations and Images for Effective Worldwide Geo-localization
ChatGPTã¯ã“ã†ä½¿ã†ã€æ·±æ´¥å¼ãƒ—ãƒãƒ³ãƒ—トã®æ¥µæ„
ChatGPTを利用ã™ã‚‹ã¨ãã«æ¼«ç„¶ã¨ãƒ—ãƒãƒ³ãƒ—トを入力ã™ã‚‹ã ã‘ã§ã¯ã€ãã®èƒ½åŠ›ã‚’å分ã«å¼•ã出ã›ãªã„。ChatGPTã®é”人ã§ã‚る深津貴之æ°ãŒã€ãƒ—ãƒãƒ³ãƒ—トã®æ¥µæ„ã‚’ä¼æŽˆã™ã‚‹ã€‚ 第5回 深津æ°ã«å¦ã¶ChatGPTプãƒãƒ³ãƒ—トã®æ¥µæ„ã€å¿…è¦ãªè¡Œå‹•ã‚’考ãˆã•ã›ã¦ã‹ã‚‰å°‹ãã‚‹ ChatGPTã®é”人ã§ã‚る深津貴之æ°ã«ãƒ—ãƒãƒ³ãƒ—トã®æ›¸ã方を整ç†ã—ã¦æ•™ãˆã¦ã‚‚らã†ã€‚今回å–り上ã’ã‚‹ã®ã¯ã€ŒReAct（Reasoning + Acting）ã€ã 。 2024.09.06 第4回 深津æ°ã«å¦ã¶ChatGPTプãƒãƒ³ãƒ—トã®æ¥µæ„ã€æ®µéšŽã”ã¨ã‚„アプãƒãƒ¼ãƒåˆ¥ã«è€ƒãˆã•ã›ã‚‹ ChatGPTã®é”人ã§ã‚る深津貴之æ°ã«ãƒ—ãƒãƒ³ãƒ—トã®æ›¸ã方を整ç†ã—ã¦æ•™ãˆã¦ã‚‚らã†ã€‚今回å–り上ã’ã‚‹ã®ã¯ã€ŒChain of Thoughtã€ã¨ã€ŒTree of Thoughtã€ã 。 2024.09.05 第3回 深津æ°ã«å¦ã¶ChatGPTプãƒãƒ³ãƒ—トã®æ¥µæ„ã€è³‡æ–™ãã®ã‚‚ã®ã‚’AIã«ä½œã£ã¦ã‚‚ら
Prompt engineering overview - Anthropic
When to prompt engineer This guide focuses on success criteria that are controllable through prompt engineering. Not every success criteria or failing eval is best solved by prompt engineering. For example, latency and cost can be sometimes more easily improved by selecting a different model. Prompt engineering is far faster than other methods of model behavior control, such as finetuning, and can
リリースã€éšœå®³æƒ…å ±ãªã©ã®ã‚µãƒ¼ãƒ“スã®ãŠçŸ¥ã‚‰ã›
最新ã®äººæ°—エントリーã®é…ä¿¡
j次ã®ãƒ–ックマーク
kå‰ã®ãƒ–ックマーク
lã‚ã¨ã§èªã‚€
eコメント一覧を開ã
oページを開ã
Agenta - Prompt Engineering, Evaluation, and Observability for LLM apps
Claude Code活用ã«ãŠã„ã¦ã€ãƒ¡ãƒ³ã‚¿ãƒ«ãƒ¢ãƒ‡ãƒ«ã‚’変ãˆã‚‹å¿…è¦ãŒã‚ã‚‹ / rethinking-claude-code-mental-models
GitHub - xai-org/grok-prompts: Prompts for our Grok chat assistant and the `@grok` bot on X.
https://x.com/rubenhssd/status/1924517497853903090?s=12&t=8snO90Ee0V5u8yV3LK7zwA
Prompt Library - Promptly AI
Conversational Prompting in Generative AI
Dynamic few-shot examples with LangSmith datasets
LLMã®è«–ç†çš„推論を強化ã™ã‚‹æ–°æŠ€è¡“「Logic-of-Thought（LoT）ã€ã€€å¾“æ¥ã®Chain-of-Thought（æ€è€ƒã®é€£éŽ–）ã®ã€Œä¸å¿ 実ãªæŽ¨è«–ã€å•é¡Œè§£æ±ºã‚’目指㙠| Ledge.ai
GitHub - dair-ai/Prompt-Engineering-Guide: 🙠Guides, papers, lecture, notebooks and resources for prompt engineering
Prompting Fundamentals and How to Apply them Effectively
{{#tags}}- {{label}}
{{/tags}}