58Design a customer churn prediction system.▼mediumAmazonNetflixMicrosoft1 replies◆ premiumA high-AUC churn model that changes no behavior is worthless. The strong answer nails the label definition, chooses classification vs survival deliberately, and is judged on retention uplift, not accuracy.Open full answer →
59Design a lead scoring system (rank sales leads by conversion likelihood).▼mediumSalesforceMicrosoftAmazon1 replies◆ premiumReps only have so many hours, so the model exists to steer them toward the leads worth calling. Interviewers listen for value-weighted prioritization, carefully built conversion labels, and evidence of revenue uplift measured against a holdout.Open full answer →