ob Overview
We are looking for a Head of Risk to build an underwriting engine that safely approves Gen Z customers often declined by traditional lenders, operating at the speed of a consumer app. You will be instrumental in replacing current decision-making processes with a robust, owned, and explainable scorecard, leveraging alternative data and on-platform behavioural signals to maximise approval rates within acceptable NPA bands. This role is critical for the company's growth, focusing on innovative risk assessment and management for a new generation of consumers.
🎯 Key Responsibilities
- Build a clean, owned, and explainable application scorecard to replace current credit decisioning processes, independent of partner NBFC overlays.
- Ingest alternative data via the Account Aggregator framework, including parsing bank statements, analysing UPI transaction patterns, merchant categories, average balance velocity, and salary credit regularity.
- Utilise device quality, PIN code vintage, and telecom data as proxies for income and stability.
- Incorporate on-platform behavioural signals such as browsing depth, Wishlist-to-purchase conversion, and time-on-app before application.
- Implement risk-tiered approvals, aiming to maximise approval rates within an acceptable NPA band.
- Build the infrastructure to capture, store, and model repayment behaviour data for the proprietary Behavioural Credit Bureau.
- Develop a feature store to make behavioural signals available to the scoring engine in real time.
- Own the model monitoring stack, including drift detection, cohort performance tracking, and vintage analysis.
- Develop the Limit Graduation Engine logic for progressive credit journeys, managing limits from small to large tickets automatically and at scale.
- Incorporate repayment timeliness scoring, purchase frequency, category migration signals, and external signals via AA refresh into the limit graduation logic.
- Implement automated limit upgrade triggers balanced against NPA guardrails.
- Develop a predictive pre-delinquency model to flag accounts likely to miss payments based on behavioural changes.
- Design and implement empathetic, digital-first early outreach strategies for pre-delinquency intervention.
- Define escalation logic that preserves brand relationships during stress events.
- Establish closed-loop feedback mechanisms to analyse intervention effectiveness across customer segments and delinquency stages.
🛠️ Required Skills
Credit Risk, Underwriting, Data Analysis, Machine Learning, Python, SQL, Account Aggregator Framework, RBI Digital Lending Guidelines, Explainable AI, Model Monitoring, Feature Engineering
✨ Good to Have
NBFC OperationsCollections StrategyBehavioural EconomicsStatistical ModellingAWS/GCP/Azure
🤝 Soft Skills
- Problem Solving
- Analytical Thinking
- Strategic Planning
- Communication
- Leadership
- Adaptability
Critical Skills
Credit Risk
Underwriting
Machine Learning
Account Aggregator Framework
🏢 About the Company
- Focus on building explainable and auditable credit models.
- Emphasis on data-driven decision making and innovation in risk assessment.
- Commitment to regulatory compliance and ethical lending practices.
- Culture of ownership and accountability in risk management.
- Iterative approach to model development and performance monitoring.
