Build Churn Prediction and Survival Models
Company: Gusto
Role: Data Scientist
Category: Machine Learning
Difficulty: medium
Interview Round: Technical Screen
Overview: This question evaluates a data scientist's competency in end-to-end churn prediction and time-to-event modeling in the Machine Learning domain, covering four model families—linear regression, logistic regression, decision-tree models, and survival analysis—while touching on problem definition, label design, feature engineering, model evaluation, interpretability, calibration, and operationalization for lifecycle-marketing. It is commonly asked to assess the ability to translate behavioral, billing, and support data into actionable, production-ready models while reasoning about class imbalance, data leakage and selection bias, relevant evaluation metrics, and the balance between conceptual understanding and practical application.