How to Build RAG Systems (in the real world)

🤝 Your team not maximizing Claude? I run 1:1 and team AI workshops for companies doing $10M+ per year: https://aibuilder.academy/yt/2peE6mwoiXs Although RAG is a powerful technique for valuable AI systems, most don't implement it properly. Here I walk through a 5-step framework for building RAG systems grounded in reality. 💻 GitHub Repo: https://github.com/ShawhinT/yt-answer-engine References [1] https://hamel.dev/blog/posts/llm-judge/#step-2-create-a-dataset [2] https://github.com/ShawhinT/yt-answer-engine/tree/main/utils/query_gen [3] arXiv:2405.07437 [cs.CL] [4] https://youtu.be/OJItZndMUII [5] https://youtu.be/982V2ituTdc [6] https://youtu.be/ayGdRbMDZcU [7] https://youtu.be/WLCbHuRr0_0 Intro - 0:00 What is RAG? - 0:16 The Problem - 2:32 Step 1: Scope the MVP - 4:06 Step 2: Create Golden Dataset - 8:05 Step 3: Build v0 Retrieval System - 15:18 Step 4: Build v0 RAG System - 19:55 Step 5: Run Experiments - 30:00 Example Code - 34:47