Program Lead - Scorecard Development

Tiger Analytics is a global AI and analytics consulting firm. With data and technology at the core of our solutions, our 4000+ tribe is solving problems that eventually impact the lives of millions globally. Our culture is modeled around expertise and respect with a team first mindset. Headquartered in Silicon Valley, you’ll find our delivery centers across the globe and offices in multiple cities across India, the US, UK, Canada, and Singapore, including a substantial remote global workforce. We’re Great Place to Work-Certified™.

Working at Tiger Analytics, you’ll be at the heart of an AI revolution. You’ll work with teams that push the boundaries of what is possible and build solutions that energize and inspire.

Work Location:

About the role:

This pivotal role focuses on the end-to-end development, implementation, and ongoing monitoring of both application and behavioral scorecards within our dynamic retail banking division. While application scorecard development will be the primary area of focus and expertise required, you have a scope of contributing to behavioral scorecard initiatives. The primary emphasis will be on our unsecured lending portfolio, including personal loans, overdrafts, and particularly credit cards. You will be instrumental in enhancing credit risk management capabilities, optimizing lending decisions, and driving profitable growth by leveraging advanced analytical techniques and robust statistical models. This role requires a deep understanding of the credit lifecycle, regulatory requirements, and the ability to translate complex data insights into actionable business strategies within the Indian banking context.

Key Responsibilities:

End-to-End Scorecard Development (Application & Behavioral):

• Lead the design, development, and validation of new application scorecards and behavioral scorecards from scratch, specifically tailored for the Indian retail banking landscape and unsecured portfolios (personal loans, credit cards) across ETB and NTB Segments. Should have prior experience in this area.
• Utilize advanced statistical methodologies and machine learning techniques, leveraging Python for data manipulation, model building, and validation.
• Ensure robust model validation, back-testing, stress testing, and scenario analysis to ascertain model robustness, stability, and predictive power, adhering to RBI guidelines and internal governance.

Cloud-Native Model Deployment & MLOps:

• Drive the deployment of developed scorecards into production environments on AWS, collaborating with engineering teams to integrate models into credit origination and decisioning systems.
• Implement and manage MLOps practices for continuous model monitoring, re-training, and version control within the AWS ecosystem.

Data Strategy & Feature Engineering:

• Proactively identify, source, and analyze diverse datasets (e.g., internal bank data, credit bureau data like CIBIL, Experian, Equifax) to derive highly predictive features for scorecard development. Should have prior experience in this area.
• Address data quality challenges, ensuring data integrity and suitability for model inputs in an Indian banking context.

Performance Monitoring & Optimization:

• Establish and maintain comprehensive model performance monitoring frameworks, including monthly/quarterly tracking of key performance indicators (KPIs) like Gini coefficient, KS statistic, and portfolio vintage analysis.
• Identify triggers for model recalibration or redevelopment based on performance degradation, regulatory changes, or evolving market dynamics.

Required Qualifications, Capabilities and Skills:

Experience:

• 3-10 years of hands-on experience in credit risk model development, with a strong focus on application scorecard development and significant exposure to behavioral scorecards, preferably within the Indian banking sector applying concepts including roll-rate analysis, swapset analysis, reject inferencing.
• Demonstrated prior experience in model development and deployment in AWS environments, understanding cloud-native MLOps principles.
• Proven track record in building and validating statistical models (e.g., logistic regression, GBDT, random forests) for credit risk.

Education:

• Bachelor's or Master's degree in a quantitative discipline such as Mathematics, Statistics, Physics, Computer Science, Financial Engineering, or a related field

Technical Skills:

• Exceptional hands-on expertise in Python (Pandas, NumPy, Scikit-learn, SciPy) for data manipulation, statistical modeling, and machine learning.
• Proficiency in SQL for data extraction and manipulation.
• Familiarity with AWS services relevant to data science and machine learning (e.g., S3, EC2, SageMaker, Lambda).
• Knowledge of SAS is a plus, but Python is the primary requirement.

Analytical & Soft Skills:

• Deep understanding of the end-to-end lifecycle of application and behavioral scorecard development, from data sourcing to deployment and monitoring.
• Strong understanding of credit risk principles, the credit lifecycle, and regulatory frameworks pertinent to Indian banking (e.g., RBI guidelines on credit risk management, model risk management).
• Excellent analytical, problem-solving, and critical thinking skills.
• Ability to communicate complex technical concepts effectively to both technical and non-technical stakeholders.