Evaluating RAG
Comes up in: AI engineering, AI system design
This concept is being written. It gets the full five steps — explanation, predict, interview angle, practice with tests, and review cards. Here is exactly what it will cover, so you can study it elsewhere in the meantime.
What it covers
- retrieval metrics (recall@k, MRR, nDCG)
- answer metrics (faithfulness, relevance, correctness)
- building a golden set
- LLM-as-judge with calibration against humans
Words you will know
What you will practise
- 50-question eval harness for the lesson RAG