Welcome to the Capstone, the final stop in the Gradient Boosting & XGBoost course. Every module ended with a guided project. This one is different: it takes a single real dataset — Adult Income — from raw data all the way to a tuned, explained, honestly evaluated model, using skills from all four modules at once.

You’ll explore and prepare the data, handle its genuine class imbalance and categorical features with XGBoost’s native tools, train with early stopping and cross-validation, tune systematically with Optuna, explain the finished model with SHAP, and report metrics honestly with the precision/recall trade-offs stated plainly. A final report traces every step back to the module that taught it.

Every number in this capstone comes from real, executed code. Start the lesson to build one complete, trustworthy model end to end.