Take Your RAG To The Max!
summary
Concrete recipe for building RAG with code and results: step‑by‑step Python examples (four worked prototypes) that use LlamaIndex to ingest web pages, emails and GitHub docs, persist embeddings in ChromaDB, and run local or OpenAI LLMs; shows chunking (1024 tokens, 20 overlap), top‑K retrieval, similarity cutoff 0.6, metadata filters, and a conversational agent via LangChain; indexes 84 conference speakers, parses PDF attachments, and returns a real dentist appointment date, so the pieces actually function end to end.
AI-written summary · about
chapters (9)
- 0:00RAG introduction
- 1:07Build RAG from scratch
- 8:18Embedding models for RAG
- 11:07Paul Graham chatbot example
- 17:33Conference speaker data indexing
- 24:47Gmail indexing example
- 30:06Documentation chatbots
- 35:51Advanced RAG techniques
- 36:11Q&A: Local vs cloud AI
related talks
41mFrom RAGs to riches: Build an AI document interrogation app in 30 mins
01h 00mKeynote: Building and Productionizing RAG
01h 12mBuild RAG using Llama 3.1
27mFrom RAGs to riches: Build an AI document interrogation app in 30 mins
18mLLMs and GenAI: A Guide to Building Your Own RAG Model for Document Interaction
48mCrafting your own RAG system: Leveraging 30+ LLMs for enhanced performance
30mI Want to Build a RAG System, Now What?
44mThe RAG Really Ties the App Together