MLOps Engineer

Evlo AI; Seattle (WA); On-site; USD 130,000 - 190,000; Full time

Job summary

Evlo AI is seeking an experienced MLOps Engineer to build and operate scalable ML infrastructure in Seattle. You will design training pipelines, deployment automation, model registries, and observability, collaborating with ML and platform teams.

This role emphasizes reliability, security, and efficient serving of real-time and batch workloads across cloud platforms.

Qualifications

• 3–8 years of experience in MLOps, ML platform engineering, or related software engineering.
• Strong Python and Linux skills with automation and deployment tooling.
• Hands-on experience with Kubernetes, Docker, CI/CD, and IaC tools like Terraform or Pulumi.
• Production cloud experience on AWS, GCP, or Azure (networking, IAM, storage, compute).
• Familiarity with ML lifecycle systems (MLflow, Kubeflow, Airflow, SageMaker, Vertex AI).
• Knowledge of model serving, feature pipelines, monitoring, and reliability principles.
• Bonus: experience with LLM inference, GPU scheduling, Ray, KServe, service meshes, or relevant technical degree.

Responsibilities

• Build and maintain end-to-end ML pipelines for training, validation, and deployment.
• Automate model packaging and release workflows with CI/CD and containerization.
• Operate model serving infra on major clouds using SageMaker, Vertex AI, or KServe.
• Implement observability for latency, throughput, drift, and cost with Prometheus, Grafana, MLflow.
• Manage experiment tracking, versioning, and artifact storage across environments.
• Improve platform reliability with IaC, testing, incident response, capacity planning, and runbooks.
• Collaborate with ML and backend engineers to optimize inference performance and safety.

Skills

Python

Linux

Kubernetes

Docker

CI/CD

Terraform

Pulumi

Cloud platforms (AWS/GCP/Azure)

Education

Bachelor's degree in CS/Engineering

Tools

Kubeflow

Airflow

GitHub Actions

GitLab CI

SageMaker

Vertex AI

KServe

Job description

About The Role

The MLOps Engineer will build and operate the infrastructure that moves machine learning models from experimentation into reliable production services. The role covers training pipelines, model registry, deployment automation, observability, and the cloud systems required to serve models at scale. Working closely with ML engineers, data scientists, and platform engineers, this role will improve release velocity without compromising reliability, security, or model quality. The team supports workloads across batch inference, real-time APIs, and LLM-enabled applications.

Key Responsibilities

• Build and maintain end-to-end ML pipelines using Kubeflow, Airflow, or equivalent orchestration frameworks for training, validation, and deployment
• Automate model packaging and release workflows with Docker, Kubernetes, Helm, and GitHub Actions or GitLab CI
• Operate model serving infrastructure on AWS, GCP, or Azure using platforms such as SageMaker, Vertex AI, KServe, or NVIDIA Triton
• Implement model and data observability for latency, throughput, drift, feature quality, cost, and production performance using tools such as Prometheus, Grafana, and MLflow
• Manage experiment tracking, model versioning, artifact storage, and promotion workflows across development, staging, and production environments
• Improve platform reliability through infrastructure as code, automated testing, incident response, capacity planning, and documented runbooks
• Partner with ML and backend engineers to optimize inference performance, GPU utilization, deployment safety, and rollback procedures

What We Are Looking For

• 3–8 years of experience in MLOps, ML platform engineering, DevOps, or a closely related software engineering discipline, including experience supporting production ML workloads
• Strong Python and Linux skills, with the ability to write maintainable automation, services, and deployment tooling
• Hands‑on experience with Kubernetes, Docker, CI/CD, and infrastructure as code using Terraform, Pulumi, or equivalent tools
• Production experience with at least one major cloud platform: AWS, GCP, or Azure, including networking, IAM, storage, and compute services
• Working knowledge of ML lifecycle systems such as MLflow, Kubeflow, Airflow, SageMaker, Vertex AI, or comparable platforms
• Understanding of model serving patterns, feature and training data pipelines, monitoring, security, and reliability engineering principles
• Bonus: experience with GPU scheduling, distributed training, LLM inference, Ray, KServe, NVIDIA Triton, service meshes, or a degree in computer science, engineering, mathematics, or a related technical field