Staff Data Scientist

The Role
The Risk Data Science team is seeking a strategic and technically deep Machine Learning Engineer to lead the evolution of SoFi’s Home Loan and Home Equity risk frameworks. This role is at the center of a high-priority initiative to deliver a comprehensive, loan-level credit risk and valuation framework for our home lending portfolios. 

As a key technical leader, you will apply state-of-the-art ML methodologies to solve complex problems—including default and prepayment modeling, property valuation (AVMs), and credit risk assessment—ensuring our high-value secured loan products are backed by world-class data science.

What You’ll Do

• Lead the design and implementation of ML models for mortgage-specific use cases, including credit underwriting, debt-to-income (DTI) validation, and automated appraisal reviews.
• Provide technical oversight and end-to-end ownership of externally developed models, ensuring they meet rigorous internal standards for quality, performance, and scalability.
• Present model performance and portfolio insights to senior leadership, translating complex data into actionable credit and business strategies.
• Collaborate with Model Risk Management (MRM) and governance teams to ensure all models are developed with the technical rigor required to satisfy complex regulatory and compliance standards.
• Partner with Product and Engineering teams to operationalize models, overseeing deployment, real-time monitoring, and seamless integration into core business systems.
• Continuously explore and leverage in-house, external, and open-source ML frameworks to build the next generation of proprietary mortgage risk tools.

What You’ll Need

• Master’s degree in Computer Science, Statistics, Mathematics, Physics, Engineering, or a quantitative field required. Ph.D. preferred.
• 5+ years of direct experience with building, implementing, and deploying machine learning models within Home Loans or Home Equity lending environments.
• Expert knowledge of statistical modeling and ML methods, including linear/logistic regression, ensemble methods (XGBoost/LightGBM), clustering, and outlier detection.
• Strong programming skills in Python and SQL.
• Hands-on experience with ML model implementation in production environments using tools such as AWS SageMaker, Git, Docker, and CI/CD pipelines.
• Strong ability to distill complex technical methodologies into simple, persuasive terms for non-technical stakeholders.

Nice To Have

• Proven track record in technical model documentation and navigating the Model Risk Management lifecycle.
• Experience working in a cross-functional capacity across Product, Engineering, and Risk departments.