Best Egg

This position is central to enhancing the end-to-end model development process, from data exploration and credit risk modeling to model deployment and monitoring in production. This role bridges the gaps between data science and MLOps and is critical to ensure that models are analytically strong, operationally reliable, and compliant. Without this backfill, we not only face delivery risk on Autonomous Under Writing (AUW), Gen6 model development but also lose key modeling bandwidth needed to maintain credit model performance and for new modeling initiatives.

Duties & Responsibilities - Strengthen Credit Risk Modeling and Credit Predictability

• Build and enhance credit risk models (e.g., Gen5B Shadow/Gen6, and Autonomous Underwriting) to improve credit predictability and manage portfolio risk.
• Work with MLOps, Feature Platform, Credit Strategy, and Implementation teams to build adaptive modeling pipelines that connect model insights directly to automated, production-ready decision flows.
• Keep our modeling framework responsive to changes in credit quality and macroeconomic trends.
• Conduct exploratory data analysis and experiment tracking to identify key risk drivers and optimize model performance.

Duties & Responsibilities - Bridge Data Science and MLOps for Reliable Model Delivery

• Own Python project structure, CI/CD setup, and end-to-end testing for reliable model delivery.
• Partner with MLOps and Feature Platform to maintain model pipelines (Metaflow, SageMaker, etc.) and streamline model deployment.
• Support model deployment and validation to make sure that models run reliably in production and deliver consistent results.
• Improve overall development efficiency through standardization, automation, version control, and reproducibility.

Duties & Responsibilities - Support Model Monitoring and Governance

• Expand monitoring frameworks to underwriting models.
• Support bank and compliance reviews of model performance, PSI/CSI analysis, reject inference, and validation.
• Maintain audit-ready documentation and data transparency for internal and external partner reviews.
• These efforts complement the broader model monitoring and compliance initiatives led by the Credit DS team.

Requirements

• Bachelor’s degree in Statistics, Mathematics, Operational Research, Computer Science/Engineering, Data Science or other quantitative major. Master's or PhD is preferred. Other quantitative fields with experience in building Cloud Services/Architect solutions, and data/CI/CD pipelines for AI/ML solutions will be considered.
• Solid understanding of the model lifecycle (EDA, modeling, evaluation, deployment)
• Intermediate Python and SQL
• Data wrangling using pandas (bonus: pyarrow, polars)
• Basic version control (git/GitHub)
• Experience with Metaflow or similar orchestration frameworks (Flyte, ZenML, SageMaker Pipelines)
• Familiarity with cloud environments (AWS preferred)
• Model deployment experience (FastAPI + Docker)
• Unit testing (pytest) and modern Python project tools (uv, poetry)
• Background in credit risk modeling or lending-related ML solutions (e.g., underwriting, early delinquency, loss prediction, fraud detection, etc.)