Reference

Compare AI Concepts

25 side-by-side comparisons of the concepts interviewers ask about most. Understand the tradeoffs, know when to use what, and never confuse the two again.

FilterAll Foundations Architecture Training RAG Inference Evaluation Production Tooling

LLM EvaluationvsRLHF (Alignment Training)

5 differences · 6 dimensions

LLM Evaluation vs RLHF

Evaluation measures quality; RLHF improves it. They are partners, not rivals, you need evaluation to even know whether RLHF worked.

Read comparison

Offline evaluationvsOnline evaluation

5 differences · 6 dimensions

Online vs Offline Evaluation

Offline eval runs against fixed datasets before deployment; online eval measures real users after deployment. You need both, and neither is a substitute for the other.

Read comparison