Prerequisites:
● Experience: 6+ years in Data Science, with at least 3 years specifically in Credit
Scoring, Risk Analytics, or Fraud Detection.
● Academic Background: Bachelor’s or Master’s or PhD in Statistics, Mathematics,
Computer Science, or Economics.
● Tech Mastery: Expert-level Python, SQL, and hands-on experience with XGBoost,
LightGBM, and SHAP for model explainability (or similar algorithms/tools)
● Domain Knowledge: Familiarity with the India Stack, Account Aggregator (AA)
frameworks, B2B business & credit cycles and lending product constructs.
Core Competencies:
● Algorithmic Architecture: Deep expertise in building PD (Probability of Default) and
LGD (Loss Given Default) models using both traditional and alternative data.
● MLOps & Engineering: Knowledge of how to move models from a Jupyter notebook
to a production-grade API that handles real-time scoring.
● Regulatory Sensitivity:



Ability to build "Glass-Box" models that comply with RBI
guidelines on transparency and bias.
● Strategic Leadership: The ability to hire and mentor a team of junior DS/DEs while
communicating risk appetite clearly to the Board and Lenders.
Success Outcomes Desired:
● At 3 Months: You have audited our current data sources and established a robust
ETL pipeline for credit-relevant features.
● At 6 Months: You have deployed our enhanced Credit Scoring Model that
outperforms traditional bureau scores by at least 15% in Gini coefficient/KS or
comparable statistics.
● At 12 Months: You have built a fully automated Early Warning System (EWS) and a
Feature Store that allows our product team to launch current risk-based products in
days, not months.

📌 Data Science Lead || Software Product || Mumbai
🏢 KSA
📍 Mumbai