MLOps/ DevOps Engineer

Job overview
We're looking for an MLOps Engineer to own the data and ML platform that powers our production scoring systems, real-time services, and cross-team data integrations. You'll manage the data platform (governance, access control, compute), build and deploy the APIs and real-time systems that serve models and data in production, and run all of it on AWS.
This role is a strong fit for someone who thinks in systems, is comfortable owning infrastructure decisions end to end, and enjoys working with both engineering teams and business/data stakeholders.
Key Responsibilities:
Data Platform Administration
• Administer the organization's data platform (e.g., a lakehouse or warehouse environment): manage access control, workspaces/environments, and compute resources
• Own data governance: catalogs/schemas, permissions, and isolation of sensitive data from general access
• Define compute and job policies to balance cost and performance
• Build and maintain CI/CD pipelines for data and infrastructure changes across dev/staging/prod
Real-Time Systems & API Deployment
• Design and deploy production APIs on AWS (e.g., API Gateway, Lambda, or containerized services) that serve models and data to internal and external consumers
• Build real-time and event-driven data flows using streaming platforms (e.g., Kafka) and low-latency stores (e.g., DynamoDB)
• Ensure these systems are scalable and observable in production: define monitoring, alerting, and incident-response runbooks, and manage the underlying cloud networking and access control needed to keep them running reliably
Model Deployment & MLOps
• Deploy, version, and monitor production model-serving APIs used by internal and external systems
• Maintain the data pipelines that feed these models, with monitoring for pipeline health and data freshness
• Coordinate production releases with minimal downtime
Data Integration & Delivery
• Design how data and predictions are delivered to different consumers: self-service query access, scheduled exports, real-time APIs, and governed sharing with external partners
• Partner with data science and business teams to choose the right integration pattern for each use case, prioritizing governed, repeatable solutions over ad hoc data transfers
Cost & Governance
• Track cloud and platform usage/cost, identify savings opportunities, and enforce tagging and budget policies