kannanjayachandran/churn-compass

A production-grade end-to-end ML system for predicting customer churn in retail banking. It uses an XGBoost model optimized for top-K targeting, with full lifecycle support including data validation, experiment tracking (MLflow), automated monitoring, drift-based retraining, and explainability via SHAP.

22
/ 100
Experimental

This project helps retail banks proactively identify customers likely to close their accounts, enabling targeted retention efforts. It takes in customer data, processes it, and generates predictions of churn risk, which are viewable on a dashboard. The primary users are customer relationship managers, marketing strategists, or risk analysts in retail banking.

Use this if you need a reliable system to predict customer churn in retail banking, with automated monitoring and a user-friendly dashboard for insights.

Not ideal if your organization is outside of retail banking or if you are looking for a simple, one-off analysis rather than a continuously operating system.

retail-banking customer-retention churn-prediction marketing-analytics risk-management
No License No Package No Dependents
Maintenance 13 / 25
Adoption 4 / 25
Maturity 5 / 25
Community 0 / 25

How are scores calculated?

Stars

7

Forks

Language

Jupyter Notebook

License

Category

mlops-end-to-end

Last pushed

Mar 22, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/mlops/kannanjayachandran/churn-compass"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.