Machine Learning Engineer, Underwriting

We're hiring a Machine Learning Engineer to build and own the models behind our underwriting and decisioning systems at FloatMe. Our models determine who gets approved, how much, and under what terms — serving customers across a wide range of profiles. The challenges are real: maintaining calibration across diverse customer populations, designing features that generalize as the portfolio grows, and balancing approval rates against loss performance at every decision point. As a senior individual contributor on our ML team, you'll work across the full modeling lifecycle — from problem formulation and feature development to deployment, monitoring, and iteration in production. We move fast, test carefully, and hold our work to a high standard because the models we build determine real outcomes for real people. If you're excited to do rigorous, high-impact ML work at a fast-moving fintech, we'd love to hear from you.

What You’ll Do

- You will be a senior individual contributor building and evolving the ML systems behind these products. You will work across the full modeling lifecycle: problem formulation, feature development, training, calibration, experimentation, deployment, monitoring, and iteration.

- Build, evaluate, and maintain underwriting and decisioning models.

- Design and evolve underwriting decision frameworks, including the modeling, automation, policy logic and amount assignment that manage exposure over time.

- Design and run experiments to evaluate model performance, measure impact on approval rates and loss, margin and inform underwriting policy decisions.

- Develop deep understanding of consumer behavior, repayment dynamics, and portfolio structure, and use that to inform model design and decision logic.

- Contribute analysis and perspective that inform portfolio-level decisions, including explaining model behavior, tradeoffs, and uncertainty to senior technical and business leaders.

- Develop and maintain the key portfolio KPIs and inventory of periodic analysis to continuously identify risk and growth opportunities

- Collaborate with Product, Engineering, Legal, Compliance, and Operations to ensure underwriting systems reflect business goals and regulatory expectations.

Technologies We Use and Teach:

- Python (NumPy, Pandas, scikit-learn, PyTorch, XGBoost, LightGBM)

- AI development tools as core infrastructure: Claude Code, Cursor, Copilot

- ML flow for experiment tracking and model registry

- Internal feature store and model hosting platform

- SQL / Snowflake

- GitHub

- AWS

- BI tools (Looker/PowerBI/Tableau)

Who You Are

- A Master degree in a quantitative field (e.g., Mathematics, Statistics, Physics, Computer Science, Operation Research). A PhD degree is strongly welcomed.

- 5+ years applying AI, machine learning, or statistical modeling in decisioning contexts such as credit, risk, fraud, recommendations, or similar domains.

- Experience with probabilistic models and decision systems, including calibration, score transformations, and interpretation of model outputs.

- Strong experimentation skills: you know how to design holdouts, measure lift, and evaluate models beyond aggregate metrics.

- Experience with model monitoring, degradation detection, and retraining strategies in production systems.

- Deep knowledge of underwriting using bank & cashflow analysis, bureau & alternative data etc. with a focus on unsecured credit risk

- Experience explaining modeling concepts, results, and limitations to senior stakeholders and cross-functional partners.

Bonus Points

- Fintech background

- Consumer finance experience (non-large bank environment)

- Advanced modeling techniques

- Background in small to medium sized companies