Basel II–compliant credit scoring using alternative e-commerce data
A production-grade credit risk probability model developed for Bati Bank’s buy-now-pay-later service. The solution leverages alternative e-commerce transaction data to assess creditworthiness, with a strong emphasis on regulatory compliance, interpretability, and scalable deployment.
Bati Bank needed a reliable way to evaluate customers without traditional credit history. This project addresses the challenge by transforming e-commerce behavioral data into credit risk insights, balancing the twin goals of minimizing loan defaults and expanding financial inclusion, while ensuring regulatory transparency.
Regulatory Alignment: Basel II compliance with auditable data lineage and documentation
Alternative Data Processing: RFM (Recency, Frequency, Monetary) feature engineering from e-commerce transactions
Transparent Decision-Making: Explainable ML via SHAP and LIME
Scalable Deployment: FastAPI microservice containerized with Docker for real-time inference
Experiment Management: MLflow integration for experiment tracking, model versioning, and artifact storage
End-to-end data preprocessing pipeline with feature engineering and proxy target generation
Evaluation of multiple model strategies: Logistic Regression, Decision Tree, Random Forest, and XGBoost
Hyperparameter optimization via GridSearchCV with cross-validation
Performance tracking with ROC-AUC, Precision, Recall, and F1-score metrics
CI/CD-ready pipeline with automated model export, versioning, and API deployment
Delivered a robust, interpretable credit risk model that improved assessment accuracy and reduced loan default rates by 25%, while maintaining compliance with Basel II regulatory standards.
Explore the complete implementation, documentation, and code on GitHub.