Sample: RAG Course Assistant

Sample content. Course knowledge was spread across slides, notebooks and video transcripts.

A retrieval-augmented assistant that answers questions from course material and cites its sources.

Chunked transcripts and notes, embedded them into a vector store, and added an evaluation loop with Opik to track faithfulness and relevance.

Faithfulness 0.87, answer relevance 0.91 on a 120-question eval set.