314,552 interview questions from 6,000+ companies.
Explain how supervised and unsupervised learning differ, and ground the distinction in a practical ML example.
A framework for connecting user needs to business goals, then making product decisions with clear trade-offs and measurable outcomes.
Explain practical strategies for handling missing values in a supervised learning workflow, from diagnosis to modeling and validation.
Explain which classification metrics to use and how metric choice depends on the business objective and error tradeoffs.
Choose hyperparameters with cross-validation and validation metrics, while balancing bias, variance, and overfitting.
Explain the bias-variance tradeoff mathematically and how L1 and L2 regularization change model complexity and weights.
Explain what drives your interest in data engineering, grounded in user needs and the value created by reliable data systems.
Build and compare baseline and engineered-feature classifiers for consumer loan default prediction, and explain how feature engineering changes model performance.
Build a loan default classifier and show how to detect and prevent overfitting using regularization, cross-validation, and model complexity control.
Build a supervised classification model to predict 12-month loan default using credit, financial, and application features.
Build a supervised model to predict client attrition risk using account activity, product usage, and support signals.
Build and compare overfit and regularized classifiers for loan default prediction, and show how cross-validation reduces generalization error.
Build and justify a credit default classifier using traditional ML, showing model selection, feature engineering, validation, and explainability.
