Address
Form of workJob summary
Develop Machine Learning based Credit risk/ Fraud risk/ other risk models and associated frameworks.Implement and monitor credit risk strategies across the lending lifecycle - acquisitions, underwriting, limit management, cross-sell, and collections.Implement and monitor fraud risk controls.
Job seniority: associate level
Responsibilities
• Deep dive into alternate data and evaluate data features for enhanced risk assessment and credit expansion.• Help in shaping the omnichannel collections strategy and use data analytics to optimize collections efficiency.• Actively coordinate with external stakeholders to ensure relevant insights are effectively utilized for credit decisioning.
Requirements
• You love (dream) data and enjoy testing hypotheses using data.• You have worked in Retail Unsecured Lending Risk Management - Credit cards, Personal loans, etc.• You know how to build ML models and scorecards and can deploy them.• You have strong data visualization skills and an analytical mind to extract insights from large and seemingly unrelated pieces.• You are enthusiastic about designing & implementing a Risk framework for a Digital Bank.• You are keen to use new technologies and data tools to deliver risk controls, design risk dashboards that are the source of truth, and keep our fingers on the pulse.• Strong fundamentals in machine learning, general statistics, and data science principles.• 2 years experience in Retail Risk management, especially in unsecured lending products.• Quantitative credit/fraud risk experience (preferably Retail Credit).• Strong SQL skills along with experience with Python.• Strong understanding of Statistical and ML techniques and experience in building risk models.• Bachelors / Masters degree in Business Management, Economics, Statistics, Finance, Computer Science, Engineering, or another quantitative field.
Key Skills Needed
• Data analysis• ML modeling• Data visualization• Risk assessment• Credit risk management• Fraud risk management• SQL• Python• Statistical techniques• Machine learning
