Skip to main content

Guided project

An enterprise RAG assistant with an eval report

Build a RAG assistant over a few hundred pages of public policy documents, with citations and refusals, and measure it with a golden question set and an eval report comparing two retrieval set-ups on accuracy, cost and speed.

You will need

  • A few hundred pages of public policy documents, such as a regulator's circulars or a state scheme's guidelines
  • Python 3, a vector store such as pgvector or a local library, and an API key

Lessons it uses

What you will show

  • A golden set of at least 30 questions with expected sources
  • An eval report comparing two retrieval set-ups on accuracy, cost and speed
  • A short demo video of the assistant answering and refusing

Sign up free to see every step

The 6 steps, the rubric to check yourself against, and ideas to go further. It is free, takes a few seconds and needs no card.

Accounts are for learners aged 18 and over. Under 18? The quick answer and the first part of every lesson stay open: browse lessons by topic.

Today's AI turns are used

Lessons, roadmaps, videos, practice cards and the code playground stay free with your account, with no limit.