ML System Design: Search Ranking and CTR Prediction

Problem Overview

An ML system design case for Applied Scientist candidates, in search, ranking and recommendation. Three candidate reports describe it: - Amazon search relevance (onsite loop, failed): how would you predict relevance for Amazon's search results? The discussion spent a long time on feature engineering, for example the cosine similarity between…

• 7 candidate-reported follow-ups — the exact probes interviewers asked, with the trigger for each
• The rest of the problem statement — full requirements, constraints, and edge cases
• Approach and trade-offs — what passing candidates did, and the mistakes that sink people