Your question is Evaluating Generative AI Performance. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
What strategies would you use to evaluate the performance of a generative AI model?
Discuss how you would combine automated metrics, human evaluation, task-specific test sets, error analysis, and production monitoring. Explain how you would handle open-ended outputs, nondeterminism, safety risks, and trade-offs between quality, latency, and cost.