Skip to content
← Back to catalog

RAG & Working with Your Own Data

by bam

Make AI answer over YOUR content (docs, policies, a product catalog, a knowledge base) without retraining a model. This is the technical, builder-track course for people who ship Retrieval-Augmented Generation (RAG). You'll build the whole pipeline: ingest and chunk your documents, embed them into vectors, store them (including in pgvector, which this app uses), retrieve the relevant pieces at question time, and assemble a grounded prompt. Then the parts that separate a demo from production: chunking strategy, retrieval quality (top-k, hybrid keyword+vector, re-ranking), context windows, and how to EVALUATE a RAG system: retrieval quality separately from generation quality, and groundedness above all. We cover the real failure modes (irrelevant chunks, stale data, prompt injection hidden inside documents), freshness and access control on your vector store, and the honest question most tutorials skip: when NOT to use RAG at all (RAG vs. fine-tuning vs. just-use-a-long-context-window). Carries the Learn.WitUS trust DNA throughout: a retrieved passage is evidence to ground and cite, never a fact to trust. F2 (Building with AI) is the recommended prerequisite; B2 (Agents & Automations) pairs well.

How this course's quizzes are builtshow

1 quiz, 11 questions in all.

  • The answer options are reshuffled on every attempt, so a retake cannot be passed by remembering where the right answer sat.
  • The quiz long enough to measure (8 or more questions) passes the length check: the right answer is as often a long option as a short one, so it cannot be picked out by length alone.
  • All 11 questions link back to the lesson that teaches the answer.
  • All 11 questions explain the answer once you have responded.

Counted from the quizzes themselves each time this page loads.

Coming soon

This course is being reviewed

Every course on Learn.WitUS is checked against its sources before it opens, so the lessons are not available yet. What the course covers is above, and any academic standards it meets are listed with it. Enrollment opens when the review is finished.

One email when it opens. Nothing else.

Want to help this course open? Test it, or vet it →

Put your name in

A course is checked against its sources before it opens. If you want to read it early and send notes, or you know this subject well enough to check the claims, say so here. The instructor decides who reviews their own course, so this is a request, not a login.

How would you like to be involved?

How we reply, and how a review invitation would reach you.

Include your country code, for example +44 20 7123 4567 or +1 555 0100.

Anything you want the instructor to know.

Your phone number and your background are shown only to this course's instructor and this school's admins. They are never published, never sold, never added to a mailing list, and never sent to any other service. Both are optional: leave them blank and email is enough.

Your name, your email address, and which of the three you picked go to the WitUS team's inbox so a person actually reads this and replies.

Already studying this course? Sign in and it will open as usual.