Production Ready RAG
summary
Claims a production-ready stack called RAGStack that ships curated, security-tested versions of LangChain, LlamaIndex, OpenAI and more and adds a continuous test harness that flags issues like a recent SQL injection in LangChain. Shows a live Next.js demo ingesting web pages with Puppeteer, chunking via LangChain’s recursive splitter, 384-dim embeddings to Astra DB, and hybrid Cohere/OpenAI calls. Cites one rollout serving ~150,000 users, a new J Vector index 74% faster and 20% more relevant, and a $25/month trial yielding ~80M queries.
AI-written summary · about
chapters (6)
- 0:00Production ready RAG intro
- 2:01Deploying generative AI apps
- 8:21RAG system components
- 13:36RAG Stack introduction
- 25:11Live RAG app demo
- 31:16Raffle for Astra DB
related talks
20mProduction RAG apps made easier with Astra
29mUsing LangChain and LlamaIndex to customize LLMs...
11mBuilding easily deployable RAG applications with Llamastack
31mBuild an Agentic RAG system using Langchain, Ollama and Milvus
01h 13mBuilding Production-Ready AI Systems in 90 Minutes
31mEasy RAG with LangChain4J and Docker
19mThe RAG Stack We Landed On After 37 Fails
01h 59mLLM applications with LangChain