Software Engineer
BUSINESS NEXT · Jul 2025 – Jul 2026
- Engineered an XGBoost model to identify at-risk customers before they leave.
- Tuned for 'Recall' to catch maximum churners, proactively saving an estimated ₹5.85Cr in customer value.
- Built a Random Forest model suggesting highly relevant loan or insurance products.
- Added exclusion rules to stop pitching already-owned products, making the sales team 2-3x more efficient.
- Created a scoring system ranking complaints by urgency, customer value, and churn risk.
- Enabling managers to resolve the most damaging issues first, reducing complaint-related issues by ~30%.
- Used K-Means clustering to group customers into 4 behavioral profiles, transitioning the business from generic marketing to highly personalized CRM campaigns.
- Built a Logistic Regression tool predicting the probability of a prospect converting.
- Prevents random calling, focusing on the highest-quality leads first to boost daily sales.


