Sr. Data Scientist - Credit Risk

Purpose Financial•Greenville, SC

3d•Onsite

About The Position

Purpose Financial is seeking a Senior Data Scientist, Credit Risk to serve as the dedicated modeling and analytics resource for our Line of Credit (LOC) products. This role owns the credit risk analytics agenda for our LOC products end to end: acquisition scoring, initial line assignment, line management, utilization and draw behavior, loss forecasting, and portfolio performance monitoring across Storefront, Digital, and Lead Generation channels. The ideal candidate has built credit models for revolving or line-based products, and can translate borrower behavior into line strategy, credit policy, and forecasted financial outcomes. This role sits within Credit Risk and Data Science, reports to the Director of Data Science & Credit Risk. and partners closely with Finance, Digital Operations, Collections, Product, and Compliance. The successful candidate is comfortable operating with a high degree of independence, moving from ambiguous business question to defensible analytical answer, and presenting that answer to senior leadership.

Requirements

Bachelor's degree in Statistics, Economics, Mathematics, Computer Science, Engineering, or a related quantitative field. Advanced degree strongly preferred.
At least three (3) to five (5) years of experience in credit risk analytics, credit modeling, or a closely related quantitative role within consumer lending.
Direct experience building and deploying credit risk models in a production lending environment is required.
Demonstrated ability to build production credit risk models using Python or R, including gradient boosting methods, logistic regression, and survival or time to event techniques.
Working fluency with the Python data science stack, including Pandas, NumPy, scikit-learn, XGBoost or LightGBM, and SHAP or comparable explainability tooling.
Strong SQL skills and the ability to work independently against large, imperfect transactional data.
Understanding of consumer credit fundamentals: probability of default, exposure at default, loss given default, roll rates, vintage analysis, and reserve or allowance concepts.
Excellent written and verbal communication skills, including the ability to present technical analysis to non-technical executive audiences in person, by phone, and through email.
Adaptability and flexibility to changing environment; and comfortable working in a dynamic, fast-paced environment. Ability to interact professionally and exhibit appropriate social skills. Ability to understand and ensure compliance with policies, procedures, and laws governing our industry, business, and products. Ability to quickly learn all technology needed to perform the role.

Nice To Haves

Experience with revolving or line-based credit products, such as lines of credit, credit cards, or open-ended installment structures, is strongly preferred.
Experience in the non-prime or subprime consumer segment is a significant advantage.
Familiarity with the regulatory environment governing consumer lending, including ECOA and Regulation B, FCRA, adverse action requirements, fair lending, and disparate impact considerations, UDAAP, and model risk management expectations consistent with SR 11-7.
Experience with reporting and visualization tools such as Tableau or Power BI.
Experience with data engineering practices, version control, and reproducible analytical workflows is a plus.

Responsibilities

Serve as the dedicated credit risk data science resource for the LOC portfolio, owning the model and analytics roadmap for the product.
Develop, validate, and maintain machine learning and statistical models across the LOC customer lifecycle, including application scoring, initial line assignment, line increase and line decrease strategy, reauthorization, and behavioral scoring.
Build and refine LOC-specific risk metric.
Design and execute champion/challenger tests and controlled experiments to optimize line assignment, fee structure, reauthorization criteria, and offer terms across customer segments and origination channels.
Produce and defend loss forecasts for the LOC portfolio, including vintage curve development, roll rate analysis, and survival or hazard-based approaches suited to open-ended revolving exposure where traditional vintage diagonals are less informative.
Partner with Finance and Product on net charge-off forecasting, net yield analysis, and budget reforecast cycles, explaining variance between forecast and actual performance in terms of mix, vintage quality, and policy change.
Recommend credit policy actions grounded in analysis, including tightening or loosening thresholds, segment-level cutoffs, and line sizing changes, and quantify the expected volume, loss, and revenue tradeoff of each action.
Evaluate alternative and bureau data sources for incremental lift in the LOC population and integrate them into production models as they become available.
Document models to standards consistent with model risk management expectations, supporting internal validation, audit, and regulatory review.
Present findings and recommendations to the Credit Risk Review Committee and other senior audiences, translating technical work into clear business implication.
Mentor junior data scientists and analysts on modeling technique, credit domain knowledge, and analytical rigor.
Understand, adhere to, and enforce all corporate policies.

Benefits

Competitive Wages
Health/Life Benefits
Health Savings Account plus Employer Seed
401(k) Savings Plan with Company Match
Paid Parental Leave
Company Paid Holidays
Paid Time Off including Volunteer Time
Tuition Reimbursement
Business Casual Environment
Rewards & Recognition Program
Employee Assistance Program
Office in downtown Greenville that offers free parking, onsite gym, free snacks/drinks