You are building a model for a consumer lending product to predict whether an applicant will default after origination. You have historical application, credit, and repayment data, and the risk team wants a model that performs well while remaining explainable enough to support business review.
How would you decide which model family is the right fit for this problem, and how would you compare candidate models before choosing one for production? In your answer, explain how you would think about features, validation, evaluation, and the tradeoffs between simpler and more complex models.