Foundation
01
🤖
What Is RAG?
Complete introduction — what, why, and how of RAG
02
💥
3 Problems RAG Solves
Private data, outdated info, hallucination — all three fixed
03
⚔️
Fine-Tuning vs RAG
When to use each — complete cost and capability comparison
04
🏗️
RAG 4-Step Architecture
Indexing → Retrieval → Augmentation → Generation
Core Components
05
📂
Document Loaders
PDF, web, CSV — load from any source into a standard format
06
✂️
Text Splitting
chunk_size, chunk_overlap, and 4 splitting strategies
07
🔄
Recursive Character Text Splitter
The production go-to splitter — how it really works
08
🧠
Embeddings
Text to vectors — how semantic meaning becomes searchable numbers
09
🗄️
Vector Stores
FAISS, ChromaDB, Pinecone — when and why to use each
10
❤️
Retrievers — Heart of RAG
Every retriever type in LangChain — complete guide
Advanced Retrievers
11
🎯
MMR Retriever
Eliminate redundant results with Maximum Marginal Relevance
12
🔍
Multi Query Retriever
Generate multiple query variants for broader retrieval coverage
13
🗜️
Contextual Compression
Extract only relevant sentences — remove all retrieval noise
Building & Production
14
🚀
Build Your First RAG System
YouTube chatbot — step by step, 30 lines of LangChain code
15
⛓️
RAG Chain Architecture
One invoke call runs the entire pipeline — chains explained
16
🔬
Advanced RAG Techniques
Production techniques — rewriting, hybrid search, re-ranking
Expert Level
17
🔧
Corrective RAG (CRAG)
Self-correcting retrieval — evaluator, refinement, web fallback
18
🪞
Self RAG
Four self-reflection checkpoints for verified, accurate answers
19
🛡️
Stopping Hallucination
Three-layer anti-hallucination strategy for production RAG
20
🎉
Series Wrap Up
Recap, cheat sheet, project ideas, and what to learn next