Data Scientist – Risk
*Experience:* 4–5 Years
*Industry:* FinTech / Digital Lending / NBFC / Banking
## About the Role
We are looking for a *Data Scientist – Risk* with 4–5 years of experience in the *FinTech/Lending industry. The ideal candidate should have strong expertise in **data analytics, machine learning, Python, SQL/PostgreSQL, and a solid understanding of **credit and lending risk*.
The candidate will work closely with Risk, Credit, Business, Product, and Collections teams to build data-driven solutions and improve lending decisions.
## Key Responsibilities
* Analyse customer, loan, repayment, transaction, and bureau data to identify *credit-risk trends and opportunities*.
* Develop and monitor *credit-risk and predictive ML models* for default, risk segmentation, collections, fraud, etc.
* Perform *portfolio, vintage, cohort, delinquency, roll-rate, and default analysis*.
* Build features, train, validate, and monitor ML models using appropriate metrics such as *AUC, Gini, KS, Precision/Recall, Lift, and PSI*.
* Write complex *SQL/PostgreSQL queries* for data extraction, analysis, and reporting.
* Use *Python (Pandas, NumPy, Scikit-learn, XGBoost/LightGBM, etc.)* for analytics and modelling.
* Create and maintain *risk/portfolio dashboards* using Metabase or other BI tools.
* Translate analytical findings into actionable recommendations for *credit policy, underwriting, risk strategy, and collections*.
## Core Skills
* *4–5 years of experience in FinTech/Lending/NBFC/Banking/Credit Risk*.
* Strong understanding of *credit risk and lending lifecycle*.
* Strong hands-on experience with *Python, SQL/PostgreSQL, Data Analytics, and Machine Learning*.
* Experience with *ML models*.
* Strong knowledge of *feature engineering, model evaluation, and statistical analysis*.
* Understanding of key lending metrics: *DPD, PAR, FPD, NPA/Default, Roll Rates, Vintage Analysis, and Recovery/Loss rates*.
* Strong analytical, problem-solving, and stakeholder communication skills.
* Exposure to *credit bureau data, fraud analytics, or collections analytics*.
* Knowledge of *model monitoring, explainability, and model/data drift*.
## Ideal Candidate
A candidate who can independently take a problem from:
*Risk/Business Problem → Data Analysis → Feature Engineering → ML Model → Validation → Business Recommendation*
and has a strong combination of *Lending/Risk domain knowledge + Data Science + SQL + Python
Pay: ₹500,000.00 - ₹1,000,000.00 per year
Work Location: In person