SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
SoFi is seeking a Fraud Model Developer to join its Fraud Model Development team. In this role, you will develop, evaluate, and monitor machine learning models that support data-driven fraud and risk decisions across SoFi's products and services, including Personal Loans, Student Loans, Credit Cards, and Crypto.
You will build quantitative and machine learning solutions designed to reduce fraud losses, minimize false positives, lower operational costs, and protect SoFi members. Key responsibilities include developing statistical and ML models using techniques such as logistic regression, gradient boosting, random forests, and neural networks; aggregating and cleaning large datasets from multiple environments; analyzing complex datasets to identify fraud patterns and loss drivers; designing, testing, validating, and recalibrating models; monitoring model performance and detecting degradation or data drift; conducting fraud-loss forecasting and scenario-based assessments; automating recurring monitoring processes and dashboards; investigating external risk data and industry trends; partnering with Engineering teams on model implementation and production deployment; and collaborating with Business Units, Operations, Product, Finance, and Risk partners to communicate findings and translate technical results into actionable business recommendations.
Required qualifications include five or more years of experience in fraud modeling, loss forecasting, advanced quantitative modeling, machine learning, or related fields. You must hold a master's or doctoral degree in Statistics, Mathematics, Economics, Engineering, Computer Science, or another quantitative field, or demonstrate equivalent professional experience. Advanced proficiency in Python and SQL is essential, along with experience creating analytical reports or dashboards using Tableau or comparable platforms. You should have hands-on knowledge of fraud-loss forecasting and fraud-reduction methodologies, experience monitoring and recalibrating models in response to performance changes, strong analytical and problem-solving skills, and the ability to work collaboratively across technical and nontechnical teams in a fast-moving environment.
