ML Platform Engineer

About this role

Hadrian seeks an ML Platform Engineer to build and operate the production infrastructure supporting autonomous factory AI systems. You'll design standardized deployment patterns around MLflow, Dagster, EKS, and FastAPI that enable reliable, secure model serving across vision, scheduling, and forecasting workloads.

What you'll do

• Design and maintain production ML serving platform with clear SLAs for latency, availability, and multi-tenancy
• Build batch and online inference systems for tabular, vision, document-AI, and embedding workloads
• Develop model release and evaluation processes including automated testing, canary deployments, and A/B tests
• Own feature serving layer, detect training-serving skew, and manage model degradation monitoring
• Create operational tooling for telemetry, incident response, autoscaling, and GPU resource management
• Build APIs, SDKs, and templates enabling model teams to deploy independently

What they're looking for

• Production Python and SQL with typing, testing, and API design
• Kubernetes, containerization, and distributed systems patterns
• ML orchestration tools (Dagster, MLflow, or similar)
• Batch and real-time inference systems
• Feature stores or feature engineering systems
• Model registries and ML CI/CD workflows
• Infrastructure cost optimization and resource scaling
• Observability and incident response