Your recommender is training on its own clicks — what breaks?
Causal inferenceHardProblem. The recommender is trained on the clicks it generates, and evaluated on logged clicks. What biases does this create, and how do you handle them in experiments and training?
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.