MLOps Engineer — Seattle, WA

About The Role

The role owns the infrastructure that makes machine learning production-ready: CI/CD for models, scalable training pipelines, low-latency serving infrastructure, and the observability layer that catches failures before clients do.

This position sits at the intersection of ML engineering and platform engineering. The MLOps Engineer will work directly with data scientists and backend teams to turn experimental notebooks into reliable, monitored systems running at scale.

Key Responsibilities

• Build and maintain end-to-end ML pipelines for training, validation, and deployment using tools like Kubeflow, ZqTGFAm, or MLflow

• Design CI/CD workflows for model versioning, automated testing, and progressive rollout (canary deployments, shadow traffic, rollback strategies)

• Operate and scale model serving infrastructure on Kubernetes, including GPU resource management, autoscaling, and latency optimization

• Implement monitoring and alerting for model performance, data drift, and infrastructure health using Prometheus, Grafana, and custom evaluation jobs

• Manage feature stores and data versioning (Feast, DVC, or equivalent) to ensure consistency between training and serving environments

• Optimize cloud costs across GPU compute, storage, and orchestration without sacrificing deployment velocity

• Document architecture decisions and runbooks; lead on-call rotation participation and incident postmortems for ML systems

What We Are Looking For

• 3–7 years of experience in MLOps, ML engineering, or DevOps with a strong ML focus

• Production experience with containerization and orchestration: Docker, Kubernetes (Helm, ArgoCD a plus)

• Hands-on expertise with at least one cloud platform: AWS, GCP, or Azure, including GPU instance management and networking

• Strong Python and scripting skills; familiarity with Terraform or similar IaC tools

• Experience with ML workflow tooling: MLflow, Kubeflow, SageMaker Pipelines, or Vertex bz XmwDS7NnX

• BS in Computer Science, Engineering, or equivalent practical experience building production ML systems

• Bonus: experience with LLM inference optimization (vLLM, TensorRT-LLM), distributed training (Ray, DeepSpeed), or GPU cost optimization at scale