Senior ML Infra Platform Engineer — Kubernetes & GPUs

Insilico Search Partners; Cambridge (MA); On-site; USD 140,000 - 210,000; Full time

Job summary

Insilico Search Partners is seeking an experienced infrastructure/ML Ops engineer to own production-grade Kubernetes clusters and data pipelines. The role focuses on scaling ML workloads, implementing CI/CD and GitOps, and ensuring reliable, observable infrastructure across on-prem and cloud.

You will collaborate with ML engineers and researchers, manage GPU workloads, and drive cross-functional platform initiatives to reduce toil and improve repeatability.

Qualifications

• 5+ years in production infrastructure, platform engineering, DevOps, SRE, or MLOps.
• Hands-on Kubernetes and infrastructure-as-code experience.
• Experience with major cloud platforms, CI/CD pipelines and containerization.
• Familiarity with ML workflows and model-serving platforms.

Responsibilities

• Manage infrastructure as code (Terraform) and production Kubernetes clusters.
• Build and maintain CI/CD and GitOps workflows; secure and optimize container environments.
• Operate workflow orchestration and storage for large ML datasets.
• Manage GPU-enabled infrastructure: provisioning, drivers, utilization.
• Establish observability (metrics, logs, dashboards, alerting) and runbooks.
• Deploy and operate ML training/inference infrastructure and tooling for experiment tracking.
• Partner with ML Engineers to move model workloads onto production-ready infrastructure.

Skills

Kubernetes

Terraform

CI/CD

GitOps

Workflow orchestration

Cloud platforms

Tools

Kubeflow

Argo CD

Nextflow

Ray

Vertex AI

Argo Workflows

Helm

CircleCI

GitHub Actions

Seqera

Job description

Insilico Search Partners is seeking an experienced infrastructure/ML Ops engineer to own production-grade Kubernetes clusters and data pipelines. The role focuses on scaling ML workloads, implementing CI/CD and GitOps, and ensuring reliable, observable infrastructure across on-prem and cloud.

You will collaborate with ML engineers and researchers, manage GPU workloads, and drive cross-functional platform initiatives to reduce toil and improve repeatability.