Building RAG Systems
Learn to ground LLMs in external knowledge using retrieval, chunking, and reranking.
Prep before this path
This path intentionally depends on a few concepts outside the track. Review these first, then use the step-level prep links as needed while you move through the path.
- Step 1
Embeddings & Semantic Search
CompletedInteractiveLearn how embeddings turn text into vectors and enable semantic search by finding meaning-based similarity instead of keyword matches.
intermediate 9 min readPrep recommended: Tokenization - Step 2
Vector Databases & Approximate Nearest Neighbors (ANN)
CompletedInteractiveLearn what vector databases store, why nearest-neighbor search must be approximate at scale, and how ANN indexes (like HNSW and IVF) make retrieval fast.
intermediate 10 min read - Step 3
Retrieval-Augmented Generation (RAG)
CompletedInteractiveLearn how RAG lets an LLM answer questions using relevant external documents fetched at query time.
intermediate 8 min read - Step 4
Chunking & Indexing Strategies for RAG
CompletedInteractiveLearn how to split documents into retrievable chunks, attach the right metadata, and index content so RAG retrieves the right context reliably.
intermediate 11 min read - Step 5
Reranking & Hybrid Retrieval
CompletedInteractiveLearn why two-stage retrieval and keyword+vector fusion improve relevance in real-world RAG systems.
intermediate 10 min read - Step 6
Context Windows & Prompt Budgeting
CompletedInteractiveBuild a practical mental model for context limits and how to allocate tokens for better cost, speed, and answer quality.
intermediate 9 min read