Senior Machine Learning Engineer

As part of SurePay’s strategic growth, we are expanding beyond our core Account Verification services to launch an integrated suite of actionable Fraud and AML risk intelligence products. To drive this, we have established a dedicated domain: Fraud Risk Intelligence.

You will join a focused cross-functional team of Engineers, Data Scientists, and Product Managers developing an enterprise-grade Machine Learning Platform.

Fraud detection requires split-second evaluation. The business goal of this role is to build, scale, and maintain a resilient, sub-millisecond real-time ML processing platform.

You will solve the critical challenge of converting experimental data science logic into high-performance, automated production systems—combining supervised risk models, unsupervised anomaly detection, and rule-based decision engines to stop financial crime in real time.

Responsibilities include:

• Product Discovery & Sparring: Partner with Data Scientists to understand the FRI problem space, discover new product possibilities, and connect technical implementations directly with client requirements and business context.
• Data Science Enablement: Support Data Scientists with their coding needs, guiding them on how to package and deploy ML models to production in a clean, safe, modular, and reproducible way (moving beyond raw notebook handoffs).
• API & Product Integration: Expose ML models to software engineers by designing high-performance microservice APIs with sub-millisecond latency, evolving new FrAML products according to planned market releases.
• Data Platform Evolution: Actively shape the next iteration of our internal data platform from an MLOps perspective, implementing real-time data streaming, aggregation points, and online feature store management.
• MLOps & Infrastructure Ownership: Architect, deploy, and maintain end-to-end AWS MLOps infrastructure using Infrastructure as Code (IaC), CI/CD pipelines, model emulation, and AWS SageMaker.
• Model Emulation & Operations: Set up automated monitoring, drift detection, continuous training loops, and model emulation environments to ensure low-latency API performance under heavy transactional loads.