Role & responsibilities

- Build and maintain ETL and feature-engineering pipelines using PySpark on Databricks
- Work on bureau + banking/alternate datasets to create credit risk and fraud features
- Develop ML based predictive models for Credit risk (e.g. Application scorecard, Behavioral Scorecard)
- Deploy models using Quick API/Flask as production APIs on cloud
- Perform model validation and monitoring using PSI, CSI, KS, GINI, AUC-ROC and decile analysis
- Communicate analytical insights and actionable recommendations to policy stakeholders.
- Support scorecard/model documentation, model governance and submission activities
- Stakeholder engagement for requirement gathering and delivery of projects
- Ad hoc analytical activities

Preferred candidate profile

- Technical Skills:



Must have Strong Python(Pandas, Scikit-Learn), SQL, PySpark + Databricks
- Must have Scorecard development Skills: WOE, IV, Logistic Regression, Reject Inferencing, Machine learning Algorithms (XGBoost, Random Forest, LightGBM etc)
- Model deployment: FastAPI or Flask
- Candidates must have min 3+ years of hands-on experience in Credit risk analytics and Scorecards development.
- Candidate must have worked in Bank/NBFC/Fintech/Credit bureaus
- Qualification: M.Sc. Statistics, B.Tech., MA Economics, MBA
- Good to have skills: Git,CICD,Tableau

📌 Credit Analyst (Pune)
🏢 SKILLVENTORY
📍 Pune