Model Driver Acceptance Probability
Company: Uber
Role: Data Scientist
Category: Machine Learning
Difficulty: medium
Interview Round: Onsite
Overview: This question evaluates a candidate's competency in production machine learning system design and operationalization, including label definition and unit of prediction, feature availability at decision time, label leakage avoidance, model selection, handling class imbalance, cold-start and delayed outcomes, evaluation and calibration, online experimentation and business metrics, and post-deployment concerns like fairness, feedback loops, and monitoring. It is commonly asked to assess the ability to apply conceptual understanding of labeling and statistical evaluation to practical deployment trade-offs; the domain is Machine Learning/Data Science and the level of abstraction spans practical application and system-level conceptual understanding.
Read the full Uber Data Scientist interview experience this question came from