Competed in the Santander Customer Transaction Prediction Kaggle challenge, a binary classification task to identify which customers will make a specific transaction in the future based on 200 anonymized numerical features.

Applied feature engineering, handled class imbalance, and benchmarked multiple models including LightGBM, XGBoost, and Gaussian Naive Bayes. Achieved strong AUC-ROC scores through hyperparameter tuning and ensemble techniques.