Role Overview
We are looking for a Fraud Analytics professional with hands-on experience in end to end model development till Validation and performance tracking. The role involves monitoring fraud risk models in production, analyzing data drift, and supporting model performance optimization using Python, PySpark, and SQL.
This is a model validation role who have work end to end in model development and understand the validation Key metrics . The focus is on model governance, validation, and fraud performance analytics.
Key Responsibilities
- validate fraud risk models in production environment
- Track model performance metrics (AUC, KS, Precision, Recall, FPR, Capture Rate)
- Conduct data drift analysis using PSI/CSI
- Identify concept drift and recommend retraining triggers
- Perform threshold analysis and strategy impact assessment
- Extract and analyze large transactional datasets using SQL and PySpark
- Build monitoring dashboards and monthly performance reports
- Work with cross-functional teams (Fraud Strategy, Risk, Model Development)
- Support audit and model validation documentation
Required Skills
Technical Skills:
- Strong SQL (complex joins, aggregations, performance optimization)
- Python (Pandas, NumPy, Sklearn model performance evaluation)
- PySpark (large-scale data handling)
- Experience calculating drift metrics (PSI, CSI)
- Knowledge of fraud KPIs and alert-based models
- Experience working on imbalanced datasets
Domain Skills:
- Understanding of fraud lifecycle (transaction fraud, card fraud, digital fraud)
- Knowledge of model monitoring frameworks
- Exposure to model governance or validation processes preferred
Good to Have
- Experience working with US/UK banking clients
- Exposure to real-time fraud models
- Understanding of regulatory expectations in model risk management
- Experience in dashboarding tools (Power BI / Tableau)
