Course

RAG & Search

Language models are brilliant generalists with a fuzzy, frozen memory — and no idea about your codebase, your docs, or last week's tickets. RAG fixes that: retrieve the right passages, then generate an answer grounded in them. This is the practical engineer's tour — embeddings, chunking, retrieval, vector databases, semantic vs hybrid search, and when to reach for RAG instead of fine-tuning. Seven short lessons, every screen drawn, no vendor lock-in.

7 lessonsIntermediate41 min

Start the course Get the app

Course lessons

What is RAG

Retrieve the right passages, then generate — so the model answers from your sources, not its fuzzy memory.

›

Embeddings

Turn text into points in space, where similar meaning sits close — the trick that lets retrieval search by meaning, not keywords.

›

The RAG pipeline

Four steps from a pile of documents to a grounded answer: chunk, embed, retrieve the top matches, and generate.

›

Vector databases

Where your embeddings live — and how a vector database finds the nearest neighbors fast enough to matter.

›

Semantic vs hybrid search

Meaning-based search is powerful but misses exact terms. Hybrid search combines it with keyword matching to get both.

›

RAG vs fine-tuning

Two ways to specialize a model — one adds knowledge at question time, the other changes behavior through training. Know which problem each solves.

›

Putting RAG to work

A quick recap of the retrieval stack — and the one principle that ties it all together.

›

Take the full course

Learn it properly — tracked & personalized

Follow the whole course as a guided journey — sequential lessons, progress, streaks and a personalized path, all in the app.

Get the app