Build a ChatGPT RAG Data Pipeline
summary
A working RAG pipeline is built that ingests a simulated X Twitter feed via Kafka into RisingWave, computes a RisingWave materialized view to extract the top 3 hashtags, and constructs prompts that query a PgVector-backed vector store. Demo runs in Docker, uses LangChain4J for embedding and retrieval, and returns relevant matches (the demo shows "multi" as the top tag). Pipeline is functional but prompt-to-RAG automation is manual in the demo and marked for automation.
AI-written summary · about
chapters (7)
- 0:00Introduction and agenda
- 1:01Quick background on generative AI
- 10:26What is Rising Wave
- 28:59Vector databases and PG Vector
- 34:34Building a RAG data pipeline
- 37:22Live demo walkthrough
- 47:11Q&A and resources
related talks
27mBuild a ChatGPT RAG Data Pipeline
49mBuild ChatGPT RAG Data Pipeline with RisingWave Stream Processor & Vector Store
32mBuilding a ChatGPT Data Pipeline with RisingWave Stream Processor and Cassandra Vector Search
45mFrom ChatGPT User to RAG Implementer: A Developer's Journey
31mEasy RAG with LangChain4J and Docker
53mBuild RAG from Scratch
59mRAG Against the Machine
41mFrom RAGs to riches: Build an AI document interrogation app in 30 mins