Data Scientist / ML Engineer — Risk & Analytics
CreditSea | Raipur, Chhattisgarh

About the role
CreditSea disburses ₹300cr+ every month across our NBFC partners and is scaling fast. Our Risk & Analytics team already owns underwriting models, collections scoring, and fraud detection on top of our in-house LOS/LMS — we're now looking to strengthen it as we scale further. We're not looking for someone who can only build models in a notebook. We're looking for someone who can ship a model into production and understand why it's making the decisions it's making.

If you can write clean code, reason about credit risk like a business owner, and don't need someone else to translate "the business problem" into "the technical problem" for you — this is your role.

What you'll do
• Join our existing Risk & Analytics team to build and ship ML models for credit underwriting, collections prioritization, and early-warning/fraud detection — using bureau data, alternate data (transaction patterns, app behavior, account aggregator data), and behavioral signals
• Own models end-to-end:



from feature engineering through production deployment, monitoring, and retraining — not just the research phase
• Raise the bar on explainability across our existing models (SHAP/LIME-level interpretability) — not as an afterthought, but because credit decisions need to be defensible to a regulator, a customer, and your own risk team
• Work directly with underwriting, collections, and product teams to understand what's actually breaking in the business, and turn that into model improvements
• Partner with data engineering to strengthen our pipelines off our LOS/LMS, payment rails, and bureau feeds

What we're looking for
• Solid coding fundamentals — you can write production-quality Python, not just prototype in a notebook
• Real ML/statistics chops — comfortable with GBMs, logistic regression, and knowing when not to reach for a deep model
• Preferred technical skill: hands-on experience with XGBoos

📌 Data Scientist (India)
🏢 CreditSea
📍 India