Nitesh K S

Mid Senior Management

Big4

Dec 2019 - Present • 6 yrs 5 mos

Data Analytics| Risk Management | Model Development Expertise in developing and validating analytical models across various risk management domains. Over a decade of experience supporting stakeholders in model development, validation, and strategic implementation for credit, operational risk, fraud detection, stress testing, and loss forecasting. Proficient in managing large-scale projects and leveraging advanced data modeling techniques to drive business value. Project & Team Management: Oversee and direct multiple analytical projects, managing extensive offshore resources to ensure high quality and timely delivery of complex models. Regulatory & Compliance Expertise: Significant experience in developing and validating models for key regulatory frameworks (e.g., CCAR, CECL), with a focus on Probability of Default (PD), Loss Given Default (LGD), and Exposure at Default (EAD) models across diverse portfolios (retail, wholesale, mortgage). Model Development: Design and implement various scorecard models for critical business functions, including application processing, collections strategies, and fraud detection systems. Emerging Risks: Active involvement in climate risk modeling and research, as well as addressing the unique risk management challenges presented by machine learning and generative AI models. Global Collaboration: Collaborate with international credit bureaus across various regions (North America, Europe, India, and the Middle East) to develop robust risk scores for retail and microfinance institutions. Technical Proficiency: Expert in data modeling and analysis using industry-standard tools and languages, including SAS (Enterprise Guide), Python, R, and Excel.

Consultant

Leading Analytics Firm

Aug 2014 - Dec 2017 • 3 yrs 4 mos

For a banking client- - Underwriting Models - The assessment of creditworthiness, ensuring that loan terms align with the applicant's risk profile. - Loan Default Prediction: Develop models (e.g., logistic regression, random forests, etc) to predict the likelihood of a borrower defaulting on a loan, using historical loan data, credit scores, income levels, and behavioral patterns. - Alternative Data Scoring: Incorporate non-traditional data sources (e.g., utility bill payments, social media activity, e-commerce behavior) to assess the creditworthiness of "thin-file" or "invisible" borrowers, thereby expanding the potential market while managing risk. Line Increase & Decrease Strategies - Dynamic Credit Limit Adjustment: Develop models that use real-time behavioral and transactional data (e.g., spending habits, payment history, account balances) to recommend optimal credit limits, increasing lines for high-value, low-risk customers and decreasing them for at-risk segments. - "Next-Best-Action" for Collections: For delinquent accounts, use predictive analytics to determine the most effective intervention strategy (e.g., proactive calls, debt restructuring offers, account freezes) to maximize recovery rates and minimize losses. - Customer Lifetime Value (CLV) Prediction: Build models to estimate a customer's CLV