Demystifying GenAI: Build a ChatGPT App with Vector Store
summary
Hands-on roadmap for building a ChatGPT-style app using real tools and patterns: PgVector as a Postgres vector store, embeddings + semantic search, and libraries for Python/JS (LangChain, Haystack) and JVM (LangChain4j, Spring AI). Includes a concrete meal planner chatbot example with JSON output format, a GitHub repo of labs, RAG retrieval flow, prompt engineering recipes, and notes on multi-agent patterns. No live full demo here; cost and scale caveats called out (one anecdote cites searches costing ~$4 each) and exercises are provided to reproduce the build.
AI-written summary · about
chapters (8)
- 0:00Introduction and survey
- 9:10AI and GenAI background
- 22:08How LLMs work
- 33:02Prompt engineering basics
- 38:40Retrieval augmented generation
- 42:52Vector databases and embeddings
- 48:25Agents and workflows
- 55:29Challenges and resources
related talks
51mEvent Driven AI: Supercharging ChatGPT with RAG & LangStream
01h 03mWhat comes after ChatGPT? Vector Databases - the Simple and powerful future of ML?
25mLets Build A Node.js ChatGPT-Like Bot With AI APIs
21mChatGPT, LLMs, and LangChains - A Beginner's Guide
27mBuild a ChatGPT RAG Data Pipeline
15mHarnessing the Power of Generative AI for Intelligent Applications
46mUsing AI in Software Design: How ChatGPT Can Help With Creating a Solution Arch.
55mAI in Java? Meet LangChain4j and Build Intelligent Services