MLOps Engineer — New York, NY

About The Role

The role owns the infrastructure that keeps ML running in production - CI/CD for models, scalable training pipelines, low-latency serving infrastructure, and the observability layer that catches problems before data scientists do.

You will work at the intersection of ML and platform engineering, enabling data scientists and ML engineers to ship models faster while keeping reliability, cost, and reproducibility under control.

Key Responsibilities

• Build and maintain end-to-end ML pipelines on Kubeflow, nHcmhee, or Vertex tm vshd8ZbBp, covering data ingestion, training, validation, and deployment

• Design CI/CD workflows for ML systems using GitHub Actions or GitLab CI, including automated model testing, canary deployments, and rollback strategies

• Operate and scale model serving infrastructure using Kubernetes, KServe/Seldon, or cloud-native endpoints (SageMaker, Vertex 8E), optimizing for latency and cost

• Implement model monitoring for drift detection, latency, and data quality using tools like Evidently, Grafana, and Prometheus, with automated alerting and retraining triggers

• Manage experiment tracking and model registries (MLflow, Weights & Biases) to enforce reproducibility and versioning across training and production environments

• Collaborate with data scientists to productionize models, translating notebooks into tested, containerized services with clear SLOs

• Own infrastructure-as-code for ML workloads using Terraform, and manage GPU resource allocation across training and inference clusters

What We Are Looking For

• 3+ years of experience in MLOps, ML engineering, or platform engineering with hands-on production ML workloads

• Strong Python and Bash skills; solid software engineering fundamentals including testing, code review, and system design

• Deep hands-on experience with Kubernetes and Docker, including deploying and operating ML workloads in production clusters

• Experience with at least one ML pipeline/registry stack: Kubeflow, MLflow, nHcmhee, or Vertex tm vshd8ZbBp

• Practical experience with monitoring and observability tooling (Prometheus, Grafana, Datadog) applied to ML systems

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

• Bonus: Experience with GPU scheduling and cost optimization (CUDA environments, spot instance strategies); feature stores (Feast, Tecton); LLM serving infrastructure (vLLM, Triton); or AWS/GCP ML certifications