2024å¹´ã®æ¯ãè¿ãã¨ãã¦ãã£ããã¨ãã¾ã¨ããã
ç®æ¨ç®¡çãScrapboxã¯è¨æ¸¬ããããã
è«æ
scholar-inboxã¨Xã§æµãã¦ããè«æããChatGPTã§è¦ç´ãããã®+æ¬æã®Figureãèªãããã«ãªã£ãã
Notionã§ç®¡çããããã«ããã11æ¬ããæ¸ãã¨ãªã£ãã
æ¸ç±
- æ¸ç±:
15
åï¼åå¹´æ¯-27å from 42åï¼
漫ç»ãèªãæéãåããªãã¾ã¾ä¸å¹´ãéããããã®ä»ã®æ¬ã¯æ¯æ10åã®èªæ¸æéã§ã³ãã³ãã¨èªãã ã ç¹ã«èªãã§è¯ãã£ãæ¬ã¯ãHARD THINGSã
æè¡æ¸
- ãã¼ã¿å¿ååææ³ âãã«ã¹ãã¼ã¿äºä¾ã«å¦ã¶å人æ å ±ä¿è·ï¼Khaled El Emam,Luk Arbuckle/ãªã©ã¤ãªã¼ã¸ã£ãã³ï¼
- ãã¼ã¿æåã¢ããªã±ã¼ã·ã§ã³ãã¶ã¤ã³ âä¿¡é ¼æ§ãæ¡å¼µæ§ãä¿å®æ§ã®é«ãåæ£ã·ã¹ãã è¨è¨ã®åçï¼Martin Kleppmann/ãªã©ã¤ãªã¼ã¸ã£ãã³ï¼
- åã·ã¹ãã å ¥é âããã°ã©ãã³ã°è¨èªã¨åã®çè«âï¼Benjamin C. Pierce/ãªã¼ã 社ï¼
ãã¸ãã¹æ¸
- ãµã¤ã¼ãªã¤ã®æ³å ãªããèªåä¸å¿ãããããã¨ããã¸ãã¹ã人çããã¾ãããã®ã?ï¼æ£å£ 泰彦/KADOKAWAï¼
- ç¬åµã¯ã²ããããªã: ãç´ äººçºæ³ãç人å®è¡ãã®æ³åï¼éåº æ¦é/æ¥çµBPãã¼ã±ãã£ã³ã°ï¼
- ããªããã¹ã±ã¼ãªã³ã°ï¼ãªã¼ãã»ãããã³,ã¯ãªã¹ã»ã¤ã§/æ¥çµBPï¼
- ãªã¼ã³ã»ã¹ã¿ã¼ãã¢ããï¼ã¨ãªãã¯ã»ãªã¼ã¹/æ¥çµBPï¼
- THE MODELï¼MarkeZine BOOKSï¼ ãã¼ã±ãã£ã³ã°ã»ã¤ã³ãµã¤ãã»ã¼ã«ã¹ã»å¶æ¥ã»ã«ã¹ã¿ãã¼ãµã¯ã»ã¹ã®å ±æ¥ããã»ã¹ï¼ç¦ç° 康é/ç¿æ³³ç¤¾ï¼
- HARD THINGSï¼ãã³ã»ããã¦ã£ãã/æ¥çµBPï¼
å°èª¬
- ä¸ä½ï¼å æ 欣/æ©å·æ¸æ¿ï¼
- ä¸ä½2 é»æ森æ ä¸ï¼å æ 欣/æ©å·æ¸æ¿ï¼
- ä¸ä½2 é»æ森æ ä¸ï¼å æ 欣/æ©å·æ¸æ¿ï¼
- ä¸ä½3 æ»ç¥æ°¸ç ä¸ï¼å æ 欣/æ©å·æ¸æ¿ï¼
- ä¸ä½3 æ»ç¥æ°¸ç ä¸ï¼å æ 欣/æ©å·æ¸æ¿ï¼
- ä¸ä½0ãã¼ãã çç¶éé»ï¼å æ 欣/æ©å·æ¸æ¿ï¼
- æé»æ (æ±æ¸å·ä¹±æ©æ庫)ï¼æ±æ¸å·ä¹±æ©/æ¥é½å æ¸åºï¼
- æªäººäºåé¢ç¸ (å°å¹´æ¢åµã»æ±æ¸å·ä¹±æ© æ庫ç 第 1å·»)ï¼æ±æ¸å·ä¹±æ©/ããã©ç¤¾ï¼
- æµ·ã«ããããä¸æåã«ï¼ç¿å ´ã¤ããï¼
è¨äº
以ä¸ã¯ä»å¹´èªãã§ããã£ãè¨äºã
æ¥åä¸ã®é½åã§ãPython/ããã³ãï¼ç¹ã«Reactï¼ã«è§¦ãããã¨ãå¤ãããã®åéã®è¨äºãèªãã§ããããæ¬åéã¯Rustãªã©ã«æ¯ã¹ã¦è¯è³ªãªè¨äºã«å½ããçãä½ããããLLMã«èããã¨ãå¤ãã£ãã
LLM/RAGã¯ãä¼ç¤¾ã®Newsletterã§ãé
ä¿¡ããã¦ãããã®+èªåã§è¦ã¤ãã¦ãããã®ï¼XãRSSï¼ãçµæ§ãªæ°èªãã ãããã®ãã¡ãè¯è³ªãªãã®ãæããã
ç¹ã«ããã£ããã®ã ã¨ããWhat We Learned from a Year of Building with LLMsãã¯ç¾å ´ã§æãåããã人ã®ç¥è¦ãå縮ããã¦ãããå
±æã¨å¦ã³ã®ä¸¡é¢ã§ã¨ã¦ãè¯ãè¨äºã ã£ãã
ã¾ããããã³ããã®æ¸ãæ¹ã§ã¯Anthropicã®ãClaude 3 Technical Diveããããã£ããAnthropicã®è¨äºã¯è¯è³ªãªãã®ãå¤ããã©ããä¸èªã®ä¾¡å¤ãããã®ã§ãRSSã«ç»é²ãã¦ããã
Python
- ãåå¿è å¿ è¦ãPythonä¸ç´è ã«ãªãããã®ãã¯ããã¯29é¸ #Python - Qiita
- pipã¨pipenvã¨poetryã®æè¡çã»æ´å²çèæ¯ã¨ãã®å±æ - Stimulator
- Python ã® __init__.py ã¨ã¯ä½ãªã®ã #Python - Qiita
ããã³ã
- React ç ä¿® (2024) - Speaker Deck
- useEffectãã¡ããã¨ç解ãã #React - Qiita
- npmã®ä¾åé¢ä¿ã«ã¤ãã¦åéããã¦ãããã¨
LLM
- LLMè¬åº§2024å¹´ãDay10. LLMã®åæã¨çè«ãï¼å¾åãã¼ãï¼ - Speaker Deck
- LLMOps : ÎMLOps - Speaker Deck
- LLM Evaluation Metrics: The Ultimate LLM Evaluation Guide - Confident AI
- AOAI Dev Day LLMã·ã¹ãã éçº Tipsé - Speaker Deck
- 大è¦æ¨¡è¨èªã¢ãã«ã«ããè¦è¦ã»è¨èªã®èå/Large Vision Language Models - Speaker Deck
- Evaluating Large Language Model (LLM) systems: Metrics, challenges, and best practices | by Jane Huang | Data Science at Microsoft | Medium
- Short Musings on AI Engineering and "Failed AI Projects"
- What We Learned from a Year of Building with LLMs (Part I) â OâReilly
- è¡æ¿ã«ãããçæAIã®é©åãªå©æ´»ç¨ã«åããæè¡æ¤è¨¼ã®ç°å¢æ´å - ãã¸ã¿ã«åº
- çæAIã®å©ç¨ã¬ã¤ãã©ã¤ã³ä½æã®ããã®æå¼ãï½ç¥ç財ç£ã»ITã»äººå·¥ç¥è½ã»ãã³ãã£ã¼ãã¸ãã¹ã®æ³å¾ç¸è«ãªããSTORIAæ³å¾äºåæã
- 社å ææ¸æ¤ç´¢&QAã·ã¹ãã ã® RAG ã§ã¯ãªãã¨ãã - Algomatic Tech Blog
- Claude 3 Technical Dive
- OpenAIã®Prompt Engineering Guideã§ããè¯ãçµæãå¾ãããã³ããã¨ã³ã¸ãã¢ãªã³ã°ãå¦ã¶
RAG
- RAGå ¥é: 精度æ¹åã®ããã®ææ³28é¸ #Python - Qiita
- RAGã®Surveyè«æããRAGé¢é£æè¡ã俯ç°ãã - å çæã®ãã¼ã¿ãµã¤ã¨ã³ãã£ã¹ãã®ã¡ã¢å¸³
- Retrieval Augmented Generation (RAG) for LLMs | Prompt Engineering Guide
- Chunking Strategies for LLM Applications | Pinecone
- RAGã®å®è£ æ¦ç¥ã¾ã¨ã #Python - Qiita
ãã®ä»
- I spent 8 hours learning Parquet. Hereâs what I discovered | by Vu Trinh | Data Engineer Things
- ãã¹ã©ã¤ãç´300æããã³ãã£ã¼ããã¼ã¸ã£ã¼ã®ããã¥ã¢ã«ï½é·æç¦åº¸@EVeM
- 令åæ代㮠API å®è£ ã®ãã¼ã¹ãã©ã¯ãã£ã¹ã¨ CSRF 対ç | blog.jxck.io
- éçºè åã MySQL å ¥é / MySQL 101 for Developers - Speaker Deck
- 人ã¯ãã¹ã©ã³ãã®æã®ã¿æé·ãã â åç°ãã
- ã¡ã«ã«ãª å°æ³ããããã®ã¨ã°ãå¦ã³ï½Shota Horii
- ã¨ã³ã·ããã¢ã¯ã¨ãã®ããã«ããã¡ã¤ã³ã«ã¿ãã¤ããã¦ãããã - Speaker Deck
Github
æ¨å¹´ã®æ¯ãè¿ãè¨äºã§ã¯private contributionãå«ãããã¦ããªãã£ãããä»å¹´ã¯å«ãããå«ããä¸ã§ï¼668â819ï¼ã+151
ã¨ãªã£ãã