An enterprise customer wants a contractual latency guarantee for our model API. What can you promise, and how?

Model latency depends on input length, output length, load and the model itself, so a naive guarantee is either meaningless or impossible. The interviewer wants a guarantee that is defined precisely, measurable and backed by capacity.

Updated Nov 2026 · Original practice questions with worked answers and teaching rubrics. Our editorial policy explains the sources, review process and limits.