Senior Applied Scientist

Role overview

Develop predictive and causal models that improve how borrowers are supported after a loan is issued. Working with product, engineering, and servicing operations, you’ll turn evidence about repayment behavior and intervention outcomes into practical decisions for payment assistance, loan changes, communications, and collections.

Responsibilities - Build and evaluate machine-learning models for repayment behavior and other servicing outcomes. - Use causal inference and experiments to measure the effects of payment assistance, loan modifications, and other interventions. - Translate predictions and estimated treatment effects into decision rules and operational actions. - Partner with product and engineering on payment flows, autopay, and borrower communications. - Work with servicing and collections teams on strategy and insights about agent performance. - Identify areas currently handled through manual optimization where machine learning can improve decisions, and explain findings and tradeoffs to technical and nontechnical partners.

Requirements - Graduate degree in a quantitative discipline such as statistics, mathematics, physics, econometrics, operations research, or computer science. - Applied experience using causal inference and experimentation to inform real-world decisions. - Strong grounding in statistics, probability, and experimental design for evaluating models or interventions. - Production-quality Python skills, including data manipulation and algorithm implementation. - At least two years conducting machine-learning, statistical modeling, or applied quantitative research in academia or industry. - Ability to frame broad business problems as research questions and communicate findings as actionable recommendations.

Nice to have - Doctorate in a quantitative field; experience with payment products such as autopay, recurring payments, or payment plans. - Familiarity with causal machine learning, treatment-effect estimation, supervised learning, feature engineering, or model evaluation on large or complex datasets. - Experience taking work from exploratory analysis and experimentation through reliable production models.

Benefits and work setup - Remote role within the United States, with regular in-person team collaboration; the cadence varies by team and is typically one or two onsite sessions per quarter.