Credit scoring system
Kuda Technologies, African neobank, 7M+ customers
50% reduction in first-payment defaults
Problem
Needed to pre-qualify millions of customers monthly for lending products. High default rates were eroding portfolio performance.
What I built
Designed and owned the credit scoring function: an optimised LightGBM model built from a 300K borrower sample, processing millions of monthly pre qualification decisions. Developed the feature stability methodology (bootstrap learning curves) that became the team's standard for all subsequent model builds.
Results
- 50% reduction in first-payment defaults.
- 75% reduction in total portfolio defaults.
- Model serves as the production scoring engine for the entire lending portfolio.
Stack
- LightGBM
- Python
- SciPy
- SHAP
- SQL
- AWS
- Bootstrap framework
Related
- Your Best Feature Might Be Your Biggest Liability (April 2026)
- Feature bootstrapping toolkit (open source)