MLOps Engineer

About The Role The MLOps Engineer owns the infrastructure and workflows that move machine learning models from research into reliable production systems. The role builds repeatable training and deployment pipelines, manages model and data versioning, and supports inference workloads across cloud and Kubernetes environments.

The team needs an engineer who can improve the operational foundation for models serving real users: faster release cycles, predictable infrastructure, clear observability, and safe rollback paths. This role works closely with ML engineers, data scientists, platform engineers, and security teams in a hybrid San Francisco environment.

Key Responsibilities

• Build and maintain CI/CD and CT pipelines for model training, validation, deployment, and rollback using tools such as GitHub Actions, Argo Workflows, and Kubeflow
• Deploy and operate model-serving workloads on AWS and Kubernetes using Docker, Helm, and Terraform, with attention to scalability, latency, and cost
• Implement model, dataset, and feature versioning workflows using MLflow, DVC, or equivalent tooling
• Create observability for training and inference systems, including metrics, logs, traces, data drift, model performance, and service-level indicators using Prometheus, Grafana, and OpenTelemetry
• Automate infrastructure provisioning, environment configuration, access controls, and secrets management across development, staging, and production
• Partner with ML and data engineering teams to standardize reproducible experiments, artifact management, release approvals, and production incident response
• Improve platform reliability through load testing, capacity planning, disaster recovery procedures, and documented operational runbooks

What We Are Looking For

• 3–8 years of experience in MLOps, platform engineering, DevOps, or production machine learning infrastructure, including ownership of deployed ML systems
• Strong Python and SQL skills, with practical experience building automation, services, and data or model lifecycle tooling
• Hands-on experience with Kubernetes, Docker, Terraform, and at least one major cloud platform, preferably AWS services such as EKS, S3, IAM, and SageMaker
• Experience implementing ML pipelines and experiment tracking with tools such as MLflow, Kubeflow, Airflow, Argo Workflows, or comparable platforms
• Working knowledge of model-serving patterns, batch and real-time inference, feature stores, model monitoring, data drift, and CI/CD for machine learning
• Bachelor’s degree in computer science, software engineering, data engineering, or a related technical field, or equivalent practical experience
• Bonus: Experience with GPU workloads, Ray, NVIDIA Triton, Feast, Databricks, service mesh technologies, SOC 2 controls, or production LLM inference systems