Data Scientist - Credit Risk

About the role

This role involves monitoring and iterating on credit risk models for Divine's undercollateralized lending system, which has already issued over one million loans. The work includes underwriting, loss forecasting, and fraud detection, with the goal of improving portfolio performance. A key aspect is developing scalable data infrastructure to track portfolio health and key risk indicators, ensuring the lending system remains robust and responsive.

The Data Scientist will collaborate with product, research, and engineering teams to define core metrics and convert them into measurable credit and growth strategies. This involves designing and analyzing experiments like A/B tests to evaluate the impact of product and policy changes. The role requires translating complex quantitative findings into clear narratives for various stakeholders, influencing decisions across the organization.

Success in this position means driving credit risk intelligence, ensuring the lending platform scales effectively while managing risk. The ideal candidate will be comfortable with ambiguity, able to work autonomously, and possess strong problem-solving skills. The work directly contributes to expanding access to credit for individuals globally who are currently underserved by traditional financial systems.

The salary for this position ranges from $135,000 to $237,000 USD.

Skills that matter here

• Python: This role requires deep proficiency in Python for end-to-end data analysis, from raw data processing to generating recommendations.
• SQL: The Data Scientist will use SQL to manage and query data, comfortable owning analyses from initial data extraction.
• Blockchain data: The role involves working with blockchain data as Divine's lending system uses stablecoins for transactions.
• Dune: Experience with Dune is relevant for analyzing blockchain data and portfolio performance within the crypto ecosystem.
• Grafana: Grafana will be used for building and maintaining monitoring infrastructure and dashboards to track portfolio health.
• Metabase: Metabase is listed as a tool for creating dashboards and tracking key risk indicators.

Who this role suits

• A person with at least four years of experience in decision science or credit risk analytics, specifically within fintech or consumer lending.
• Someone who understands credit risk modeling concepts like PD/LGD modeling, scorecard development, and risk segmentation.
• A professional who can influence cross-functional teams and senior stakeholders by translating analytical insights into actionable strategies.
• An individual comfortable with ambiguity, possessing a bias toward action, and capable of working with minimal oversight.

From the employer

• Monitor credit risk models, including underwriting, loss forecasting, and fraud detection, and iterate based on observed portfolio performance.
• Design, build, and maintain scalable data pipelines, monitoring infrastructure, and dashboards to track portfolio health, user behavior, and key risk indicators.
• Partner with product, research, and engineering teams to define north star metrics and translate them into measurable, actionable credit and growth strategies.
• Design and analyze A/B tests, quasi-experiments, and causal inference studies to evaluate the impact of product and policy changes.
• Produce portfolio monitoring and investigative analyses, making recommendations based on findings.
• Translate complex quantitative findings into clear, compelling narratives for product, leadership, and cross-functional stakeholders.

• 4+ years of experience in decision science, credit risk analytics, or a closely related quantitative role within fintech or consumer lending.
• Deep proficiency in Python and SQL; comfortable owning analyses end-to-end from raw data to recommendation.
• Strong understanding of credit risk modeling concepts, including PD/LGD modeling, scorecard development, reject inference, vintage analysis, and risk segmentation.
• Demonstrated experience monitoring credit risk metrics and portfolio performance, including loss forecasting and underwriting model improvement.
• Proven ability to influence and collaborate with cross-functional teams and senior stakeholders, with a track record of translating analytical findings into accessible, actionable insights.
• Experience designing and evaluating experiments (A/B tests, holdout groups, or causal inference frameworks) in a consumer product context.
• Comfortable with ambiguity and biased toward action; thrives with minimal oversight and brings strong problem-solving skills and sharp attention to detail.