Naive RAG

The foundational RAG pattern: retrieve relevant chunks, pass them to an LLM, get a grounded answer. Simple, effective, and the baseline everything else builds on.

Prerequisites: Complete Foundations first.

Module	Topic
01 - Retrieval	Semantic search, top-k, similarity scores
02 - Generation	Prompt design, faithfulness, full Q&A loop
03 - Cloud Models	Gemini and other APIs, model-agnostic design