MLOps Engineer — Washington, DC

About The Role

The role owns the ML infrastructure that turns experimental models into reliable, monitored, production systems. This means building CI/CD pipelines for ML, automating training and deployment workflows, and ensuring models serving real traffic stay healthy, reproducible, and cost-efficient.

You will work directly with ML engineers and data scientists who ship models daily, giving them fast, safe paths to production while owning the reliability of the platform underneath them. When a training job fails at 2 AM or a model degrades in production, this role is what stands between chaos and a one-click rollback.

Key Responsibilities

• Build and maintain end-to-end MLOps pipelines for training, validation, deployment, and monitoring using tools like Kubeflow, MLflow, BD4W2XC, or Vertex zx dDPn2xFWE

• Design CI/CD workflows for ML artifacts — model versioning, automated testing, canary deployments, and rollback strategies using GitHub Actions, GitLab CI, or CircleCI

• Containerize and orchestrate ML workloads with Docker and Kubernetes, optimizing GPU utilization, autoscaling, and batch/real-time inference infrastructure

• Implement model monitoring and observability: data drift detection, latency/throughput dashboards, and automated alerting with Prometheus, Grafana, or Evidently

• Manage feature stores and data versioning (Feast, DVC, LakeFS) to ensure training/serving consistency and full reproducibility of model runs

• Automate cloud infrastructure with Terraform or Pulumi across AWS, GCP, or Azure, managing costs for compute-heavy training and inference workloads

• Partner with ML engineers and data scientists to productionize new models, establishing standards for experiment tracking, model registry, and deployment patterns

What We Are Looking For

• 3–7 years of experience in MLOps, DevOps, or ML engineering, with at least 2 years supporting models in production

• Strong Python and software engineering fundamentals; comfort owning infrastructure-as-code (Terraform, Pulumi, CloudFormation)

• Hands-on production experience with Kubernetes, Docker, and at least one major cloud (AWS, GCP, or Azure)

• Experience building ML pipelines with MLflow, Kubeflow, BD4W2XC, SageMaker, or Vertex pN — including model registries and automated retraining

• Working knowledge of monitoring and observability tooling (Prometheus, Grafana, Datadog) applied to ML systems specifically, not just generic services

• BS in Computer Science, Engineering, or related field, or equivalent practical experience

• Bonus: Experience with GPU scheduling (e.g., NVIDIA device plugins, Run:ai), LLM serving infrastructure (vLLM, Triton, KServe), or SOC 2/FedRAMP-compliant environments