Lead AI-Native Products with Microsoft's Agentic AI Program
AI, Data Science & Cloud Certificates from Google, IBM & Meta
Overview
Build a Learning Habit
Download Class Central's free printable study calendar
Download for Free
Learn to build a production-ready Retrieval-Augmented Generation (RAG) pipeline using LangChain that addresses common hallucination issues through advanced retrieval techniques. Discover why naive RAG systems fail and implement hybrid search combining BM25 with semantic search for improved document retrieval. Master re-ranking techniques using ColBERT to enhance precision in your results. Explore HyDE (Hypothetical Document Embeddings) query enhancement methods to generate better search queries. Build a complete RAG retrieval pipeline with proper citations and source code examples. Gain practical experience with production-grade techniques including keyword-based search integration, neural re-ranking models, and query transformation strategies that significantly improve retrieval accuracy and reduce AI hallucinations in real-world applications.
Syllabus
00:00 - Why naive RAG fails
04:04 - BM25 and hybrid search
07:03 - Re-ranking with ColBERT for precision
08:38 - HyDE query enhancement
10:32 - Full RAG retrieval pipeline with citations
Taught by
Venelin Valkov