Experience Level
Work Experience
Education
Qualifications
Industry Experience
I am excited to share the results of my latest project, which is an end-to-end Machine Learning pipeline engineered to automate and optimize credit risk assessment.
Manual credit scoring is often slow, vulnerable to human bias, and struggles with non-linear financial indicators. To solve this, I developed a data-driven framework using an advanced XGBoost Classifier to evaluate borrower risk dynamically.
Key technical highlights from the project:
Engineered Impactful Features: Synthesized key metrics like Debt-to-Income (DTI), Loan-to-Income, and Experience ratios to capture true repayment headroom.
Class Imbalance & Leakage Protection:
Integrated a robust ColumnTransformer preprocessing pipeline alongside algorithmic weight scaling to ensure clean out-of-sample generalization.
Risk-Averse Threshold Tuning:
Shifted the classification boundary to a strict 70% confidence cutoff, significantly reducing False Positives and protecting institutional portfolio capital.
Hire a Data Scientist
We have the best data scientist experts on Twine. Hire a data scientist in Nairobi today.