Kumar Prateek Swain

About Kumar Prateek Swain

I am a Data Scientist with experience across credit risk, predictive modeling, and end-to-end ML pipelines. My journey spans working at ExxonMobil (Cost Engineering) and transitioning into advanced analytics & risk modeling at a fintech. I specialize in building scorecards, designing XGBoost, Decision Tree and logistic regression models, and creating scalable pipelines for risk prediction. Beyond work, I contribute to open-source projects on GitHub (ML pipelines, credit risk models, data engineering utilities). I am passionate about applying data science in finance, risk management, and regulatory contexts (RBI, SEBI, NBFCs).

Experience

1. Consultant

   Mathlogic

   Sep 2024 — Present · Gurugram, IN

   Developed credit risk models using logistic regression and GLM to predict loan defaults, improving KS/AUC metrics by 15%.• Automated data pipeline and feature engineering with Python (pandas, SQL) for reproducible model training and evaluation.• Built a Python package for KS/Gini metric evaluation and dashboards, reducing analysis time and standardizing model validation.

Education

• National Institute of Technology Rourkela

  Bachelor of Technology - BTech

  2018 — 2022