Staff Data Scientist

Earnest•San Francisco, CA

3d•$216,000 - $262,000

About The Position

The Staff Data Scientist will own risk data science at Earnest, developing credit risk, underwriting, fraud detection, collections, and loss forecasting frameworks that shape portfolio health and the unit economics of lending products. The role leads high-impact modeling initiatives, mentors data scientists, and partners with senior stakeholders across Risk, Product, Capital Markets, Engineering, Finance, and Compliance. The position is responsible for end-to-end development and delivery of risk science systems, including credit risk scoring, default prediction (PD/LGD), fraud detection, and collection scores. It leads model design, backtesting, validation, and loss forecasting; serves as an internal subject matter expert on risk modeling and risk-based decisioning; and drives cross-functional projects through production deployment. The role also communicates analyses to senior leadership and the Risk Committee, supports modeling standards and governance, and oversees scalable risk and fraud pipelines, monitoring, stress testing, and alerting. It partners with the MRM team on model validation, fair lending analysis, governance, and technical documentation. The role is based in the US and reports to the Sr. Manager, Data Science + Pricing.

Requirements

At least 8 years of experience with Python or R and SQL.
At least 7 years of professional experience in quantitative risk modeling, credit risk analytics, fraud analytics, or applied data science.
Master's or PhD in Economics, Statistics, Computer Science, Operations Research, or a related quantitative field.
Prior experience in consumer lending, fintech, or financial services, with direct technical ownership of credit underwriting or fraud models.
Strong expertise in credit risk modeling, default prediction (PD, LGD, EAD), survival analysis, machine learning algorithms such as Gradient Boosted Trees and Logistic Regression, and risk-adjusted return frameworks.
Deep proficiency in model lifecycle management, model validation, backtesting, and monitoring in credit or fraud risk environments.
Proven ability to independently own and deliver complex, cross-functional modeling initiatives end-to-end.
Experience mentoring or managing junior data scientists and contributing to team-wide technical and governance standards.
Strong communication skills and ability to influence stakeholders and clearly present technical concepts to senior audiences.

Nice To Haves

Understanding of financial metrics and consumer credit regulations such as FCRA and ECOA/Fair Lending.
Experience with cloud production platforms such as AWS, GCP, or Azure.
Proficiency with modern data engineering and analytics tools such as Looker, Airflow, and dbt.
Experience using non-traditional data sources or LLM/NLP tools for underwriting, fraud detection, or document processing.

Responsibilities

Own end-to-end development and delivery of risk science systems, including credit risk scoring, default prediction (PD/LGD), fraud detection, and collection scores.
Lead design, backtesting, and validation of credit risk models and loss forecasting frameworks using statistical techniques and machine learning.
Serve as the internal subject matter expert on credit risk modeling, loss dynamics, risk-based decisioning, credit policy, and risk management.
Lead high-visibility, cross-functional risk modeling projects from conception through production deployment, managing timelines and stakeholder expectations.
Explain risk models, loss projections, and analytical findings to senior leadership, the Risk Committee, and non-technical stakeholders.
Mentor data scientists and provide technical guidance, helping establish modeling standards, governance, and analytical rigor.
Partner with Risk, Pricing, Product, Capital Markets, Engineering, and Compliance leadership to align risk strategies with company growth, funding requirements, and regulatory obligations.
Oversee scalable, production-grade risk and fraud pipelines with data and ML engineering teams, including monitoring, stress testing, and alerting.
Partner with the MRM team on model validation, fair lending analysis, model governance, and technical documentation.