Atoms
Staff Cluster Infrastructure Engineer
$224K–$284K
San Francisco, CA +1 · Staff · Full-time · Posted Jun 26, 2026
- Location
- San Francisco, CA
San Francisco, California, United States - Experience
- 6+ years operating GPU compute on Kubernetes or similar orchestration.
About the job
Own and scale the GPU compute infrastructure used to train Atoms’ foundation models, automating cluster management and supporting reliable operations.
What you'll do
- Manage and automate GPU training clusters, including provisioning, bare-metal bring-up, and lifecycle management.
- Build software abstractions for training and simulation workloads, and improve automation and uptime at the hardware/software boundary.
- Operate and troubleshoot infrastructure, designing it to scale from a small cluster to a larger fleet.
What they ask for
- 6+ years of experience operating GPU compute on Kubernetes or similar orchestration.
- Strong programming and scripting skills in Python, Go, or similar.
- Familiarity with Infrastructure-as-Code tools such as Terraform or CloudFormation.
- Comfort with bare-metal Linux environments, GPU hardware, and networking.
Pay
$224,000 - $284,000 per year (base salary)
Tools
GPU, Kubernetes, Python, Go, Terraform, CloudFormation, Linux
AI focus
Operate GPU infrastructure that trains foundation models and supports training and simulation workloads.
Model-written summary · the original posting is authoritative