Senior Platform Engineer: AI Compute & GPU on Kubernetes

Fdetimes; San Diego (CA); On-site; USD 190,000 - 230,000; Full time

Job summary

Shield AI in California is seeking a Senior Staff Platform Engineer to build and operate the AI Factory compute platform. You will enable GPU workloads on Kubernetes, design resource quotas and isolation, and automate deployment of GPU services across cloud and on‑prem environments.

You’ll partner with ML engineers to diagnose issues, improve throughput, and guide capacity planning with hardware recommendations.

Qualifications

• Experience building production Kubernetes platforms with GPU workloads.
• Strong software engineering in Python, Go, or similar.
• Experience automating infrastructure deployment and upgrades.
• Practical Linux, networking, and storage troubleshooting.
• Ability to work with ML engineers and translate workloads into platform capabilities.

Responsibilities

• Build and maintain software and configuration for GPU workloads on Kubernetes.
• Design quotas, scheduling priorities, permissions and isolation to share compute.
• Automate deployment and upgrades of GPU services like Run:ai using IaC.
• Investigate why jobs wait, fail or run slowly and fix related issues.
• Build diagnostics linking queue times, GPU utilization and throughput to experience.
• Benchmark workloads and advise on capacity and hardware needs.

Skills

Kubernetes

Python

Go

Infra automation

Linux

Networking

Storage

Scheduling policies

Collaboration with ML engineers

Tools

Terraform

Helm

Run:ai

Job description

Shield AI in California is seeking a Senior Staff Platform Engineer to build and operate the AI Factory compute platform. You will enable GPU workloads on Kubernetes, design resource quotas and isolation, and automate deployment of GPU services across cloud and on‑prem environments.

You’ll partner with ML engineers to diagnose issues, improve throughput, and guide capacity planning with hardware recommendations.