Production-grade end-to-end Machine Learning and AI portfolio projects featuring complete data pipelines, modeling, XAI, and cloud deployment.
Production-grade end-to-end regression portfolio project: Ames Housing dataset, log target normalization, ColumnTransformer pipelines, XGBoost/Random Forest GridSearchCV tuning, SHAP explainability, and FastAPI/Docker deployment.
Production-grade financial anomaly detection portfolio project: Highly imbalanced transaction data (0.17% fraud), SMOTE resampling inside cross-validation, XGBoost classifier, Recall/PR-AUC optimization, and SHAP feature attribution.