Using LangChain and LlamaIndex to customize LLMs...
summary
Demos show a working RAG pipeline: ingest a 15‑page Sherlock Holmes PDF, chunk and embed text, store vectors (FAISS/Chroma), and answer queries with a LangChain-powered generator that uses an Octo AI Llama 2 7B endpoint. Code is in a public repo and also crawls Octo AI and Kubernetes docs to build a web retriever, comparing responses from 27b vs 70b models (70b is notably more concise). Deployment uses containerized AWS Lambda via SAM and the talk demonstrates LangChain + LlamaIndex integration end‑to‑end.
AI-written summary · about
chapters (8)
- 0:00Introduction and agenda
- 2:21Octo AI platform overview
- 6:54What is RAG and its components
- 10:06Demo 1: Basic LLM interaction
- 14:24Demo 2: RAG on PDF document
- 17:10Demo 3: Multi-model comparison
- 20:49Deploying to AWS Lambda
- 27:29Recap and Q&A
related talks
27mAI Demystified Unveiling the Intelligence Behind the Machine
01h 59mLLM applications with LangChain
42mUsing LlamaIndex for building Generative AI apps
55mBuild enterprise generative AI apps using Llama 3 at 1,000 tokens/s on the SambaNova AI platform
31mBuild an Agentic RAG system using Langchain, Ollama and Milvus
18mLLMs and GenAI: A Guide to Building Your Own RAG Model for Document Interaction
36mProduction Ready RAG
27mBuilding Useful Apps With Small, Local LLMs