Your question is Decision Tree From Scratch. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
How would you implement a decision tree algorithm from scratch?
Implement a practical CART-style decision tree without using an existing tree estimator. Support binary classification with numerical features, explain the impurity calculation and split-search strategy, and include stopping criteria such as maximum depth and minimum samples per leaf. Your solution should demonstrate training, prediction, and evaluation on a clearly defined train/test split. Discuss how you would extend the implementation to regression, categorical features, missing values, pruning, and class imbalance.