A new classifier shipped. Is the platform safer?
Measure successMediumProblem. The ML team shipped a much better enforcement classifier for one policy. Their offline evaluation shows a big recall improvement. They want to declare the platform safer. How would you define success?
Before you reveal: say your answer out loud, as if you were in the real interview — get your reasoning across clearly first. There is no single correct answer: reading what the interviewer is really after and defending your own thinking is what makes an answer strong.