Inherited Model Evaluation

Let’s say you’ve just joined a team at a fintech startup, and you’re now responsible for maintaining a credit risk model that predicts loan defaults. The model was built by a previous data scientist who has since left, and you have access to their code, documentation, and a sample of recent model outputs.

What steps would you take to evaluate and take ownership of this model, and how would you decide what (if anything) to change first?