Credit Risk Data Scientist

Job Summary (List Format) – Credit Risk Data Scientist

• Design, build, evaluate, and defend machine learning models to predict credit risk for short-term lending products (e.g., tax refund advances, BNPL, installment loans).
• Collaborate with product and risk teams to align models with business goals and drive actionable lending decisions.
• Develop efficient and reusable data pipelines for feature generation, model development, scoring, and reporting using Python, SQL, and CK ML/AI infrastructure.
• Deploy models to production in collaboration with data science and engineering teams.
• Partner with business stakeholders to create policies based on model results.
• Implement and monitor model performance metrics (AUC, KS, Gini) and stability indices (PSI, CSI).
• Ensure model fairness, interpretability, and compliance with relevant regulatory frameworks (FCRA, ECOA).
• Utilize expertise in Python (scikit-learn, XGBoost, LightGBM, pandas, numpy) and SQL for large-scale data processing.
• Apply advanced machine learning techniques (tree-based models, regression, time series, causal analysis, clustering) to solve credit risk problems.
• Analyze and interpret large, complex datasets to extract actionable insights.
• Leverage deep understanding of credit risk modeling concepts (PD calibration, reject inference, adverse action logic, risk segmentation).
• Work with tax and/or credit bureau data (e.g., TransUnion, Experian, Equifax) in credit model development.
• Incorporate cash flow data and other alternative data sources into credit risk models.
• Demonstrate strong business problem-solving, communication, and collaboration skills.
• Requires a degree in Mathematics, Statistics, Computer Science, or related field, and at least 2+ years’ experience in Data Science, Machine Learning, and credit risk/lending/fintech domains.
• Overall 9+ years of relevant professional experience required.