Staff Machine Learning Engineer, Credit Products (Square Financial Services)

Details

• Apply a rigorous scientific mindset to the challenge of underwriting new customer segments, involving the evaluation of alternative external data sources and the deployment of advanced architectures to enhance predictive accuracy.
• Lead complex ML Operations and Infrastructure initiatives that advance our modeling capabilities, such as scaling data ingestion or enabling the use of more complex neural networks.
• Design and implement the full credit modeling stack, taking responsibility for the entire lifecycle of credit decisioning and ensuring models are robustly integrated into production environments.
• Use data science techniques to leverage new data sources for modeling, making sense of messy datasets and bringing clarity to business decisions.
• Identify and execute material improvements to credit policy, applying an analytical lens to determine where technical or logic shifts can yield significant positive outcomes for the customer and the bank’s portfolio.
• Support team members in ad-hoc and scheduled updates to existing models, and help troubleshoot issues in a real-time production environment.
• Operate effectively within the framework of a regulated bank (SFS), balancing rapid innovation with the requirements of safety, soundness, and compliance.
• Minimum of 8 years of related experience with a Bachelor's degree; or 6 years and a Master's degree; or a PhD with 3 years experience, with a focus on developing and deploying machine learning and statistical models in production environments.
• A degree in a technical field (e.g., Computer Science, Mathematics, Statistics, Physics, or Engineering). We have a strong preference for candidates with a demonstrated track record of scientific research or an advanced degree.
• Strong quantitative intuition and data visualization skills, with a proven ability to conduct sophisticated ad-hoc and exploratory analysis.
• Full-stack proficiency preferred, including the ability to contribute across the entire technical stack—from data pipelines to production-grade software architecture.
• The versatility to communicate clearly with both technical and non-technical audiences, particularly in the context of high-visibility projects and executive stakeholders.
• A pragmatic approach to problem-solving, with a willingness to utilize whichever tool is most appropriate for the situation while balancing complex business, technical, and regulatory constraints.
• Experience with tree-based models and gradient boosting is helpful but not required; we value the ability to adapt and learn new methodologies as the credit landscape evolves.