GPU AI Platform Engineer | Kubernetes & DevOps Leader

Socket.dev

San Jose, Northern (CA, KY)

Hybrid

USD 140,000 - 200,000

Full time

3 days ago

Be an early applicant

Get more replies from employers

Send a job-specific resume in minutes.

Job summary

AMD is seeking a DevOps / Platform Engineer in San Jose, CA to build and operate large-scale GPU compute infrastructure powering AI and ML workloads. The role demands leadership, effective communication, and willingness to contribute across startup-like needs within a large company.

You will design and extend platform capabilities, manage Kubernetes deployments with Helm and GitOps (ArgoCD/Flux), and collaborate with product teams to deliver robust developer platforms and scalable

Qualifications

Experience in DevOps, Platform, or Infrastructure Engineering.
Deep hands-on experience with Kubernetes and container orchestration at scale.
Proven ability to design and deliver platform features that serve internal customers or developer teams.
Experience building developer-facing platforms or internal developer portals.
Hands-on experience in storage or network engineering within Kubernetes environments (CSI drivers, dynamic provisioning, CNI plugins, or network policy).
Experience with Infrastructure as Code tools like Terraform.
Background in HPC, Slurm, or GPU-based compute systems for ML/AI workloads.
Practical experience with monitoring and observability tools (Prometheus, Grafana, Loki).
Understanding of machine learning frameworks (PyTorch, vLLM, SGLang).

Responsibilities

Build and extend platform capabilities to enable new workloads.
Design and operate scalable orchestration systems using Kubernetes across on-prem and multi-cloud.
Develop platform features such as secret management, configuration management, deployment automation.
Partner with development teams to extend GPU developer platform with features, APIs, templates, and self-service workflows.
Manage service lifecycle within Kubernetes using Helm and GitOps workflows (ArgoCD or Flux).
Apply expertise in storage and networking to design CSI drivers, persistent volumes, and network policies.