Going Beyond Prompt Engineering
summary
Shows how to turn LLM calls into real apps by adding retrieval and tooling: implements retrieval-augmented generation (RAG) with a vector database and embeddings, uses LangChain + OpenAI embeddings in a demo web app that loads documents, chunks them, stores vectors and answers queries. Demo runs in about 25 lines of frontend code vs ~100 lines for a backend RAG flow, uploads a ~28,000-line dataset for indexing, and demonstrates in-memory vectors working; also outlines fine-tuning and agent-based tool use.
AI-written summary · about
chapters (9)
- 0:00Introduction to AI beyond prompt engineering
- 3:56Evolution of AI models and context awareness
- 17:01Basics of prompt engineering
- 24:04Making LLMs context aware with RAG
- 35:39Building RAG applications with LangChain
- 40:40Simplifying AI apps with Flows Engine
- 41:43Fine-tuning vs RAG vs long context
- 43:29AI agents and autonomous systems
- 47:20Summary and future of AI applications
related talks
50mRetrieval Augmented Generation Beyond Toy Demos
01h 28mFrom RAG to Riches Building Enterprise Search Systems with LLMs
01h 00mLLMs gone wild
41mFrom RAGs to riches: Build an AI document interrogation app in 30 mins
57mLLMs gone wild
01h 33mRAG How To Talk To Your Data
28mAPIs and GenAI : Your Perfect Combo with LLM-Powered Frameworks.
48mCrafting your own RAG system: Leveraging 30+ LLMs for enhanced performance