What are the responsibilities and job description for the MLOps Engineer position at Evlo AI?
About The Role
The role owns the infrastructure that turns experimental ML code into reliable, scalable production systems - CI/CD for models, GPU compute orchestration, feature stores, and the observability stack that keeps dozens of models healthy in production.
This is a high-impact role: ML engineers ship faster and data scientists iterate more when the platform underneath them works. The role will build tooling and pipelines that the entire AI organization depends on daily.
Key Responsibilities
The role owns the infrastructure that turns experimental ML code into reliable, scalable production systems - CI/CD for models, GPU compute orchestration, feature stores, and the observability stack that keeps dozens of models healthy in production.
This is a high-impact role: ML engineers ship faster and data scientists iterate more when the platform underneath them works. The role will build tooling and pipelines that the entire AI organization depends on daily.
Key Responsibilities
- Build and maintain CI/CD pipelines for ML workloads using tools like GitHub Actions, GitLab CI, or Argo Workflows, covering model training, evaluation, and deployment stages
- Design and operate Kubernetes-based model serving infrastructure (KServe, Seldon, Ray Serve, or custom), with autoscaling strategies for both CPU and GPU workloads
- Instrument end-to-end monitoring for production models: data drift detection, latency/throughput SLOs, and automated rollback triggers using Prometheus, Grafana, and custom alerting
- Manage ML experimentation and lineage tooling (MLflow, Weights & Biases, or equivalent), ensuring reproducibility across training runs, artifacts, and deployments
- Optimize cloud costs and utilization across GPU fleets - spot instance strategies, batch scheduling, and right-sizing for training and inference workloads on AWS, GCP, or Azure
- Build self-service tooling that lets data scientists deploy models to production without platform team intervention, including templates, golden paths, and internal documentation
- Partner with ML engineers and data scientists on architecture reviews, incident response, and postmortems for production ML systems
- 3–7 years of experience in MLOps, DevOps, or infrastructure engineering, with at least 2 years supporting ML systems in production
- Strong proficiency in Python and/or Go, plus deep hands-on experience with Kubernetes in production environments
- Experience with major ML platforms and tooling: MLflow, Kubeflow, Vertex AI, SageMaker, or equivalent model lifecycle systems
- Solid grasp of infrastructure-as-code (Terraform) and cloud networking, IAM, and security best practices on at least one major cloud
- Experience building observability for distributed systems: metrics, logging, tracing, and alerting at scale
- BS in Computer Science, Engineering, or equivalent practical experience
- Bonus: Experience with GPU scheduling and multi-tenancy (Ray, Volcano, Run:ai), feature store implementations (Feast, Tecton), LLM inference infrastructure (vLLM, Triton), or SOC 2 / compliance-driven infrastructure work