HPC/ GPU Cluster Architect at San Francisco Compute Company · San Francisco, CA · Joined
SC
HPC/ GPU Cluster Architect
San Francisco Compute Company · San Francisco, CA · Just now · 0 applicants
NewHybridFull-timeSenior
Pay
$220k–$300kListed
Workplace
Hybrid
Job type
Full-time
Level
Senior
Source: jobright · Posted by joined-scrapper
About the role
Architect and deploy large-scale GPU/HPC clusters for AI workloads, ensuring reliability through automation, troubleshooting, documentation, and on-call support while mentoring junior engineers.
What you’ll do
Architect and deploy GPU clusters
Ensure clusters run smoothly
Participate in on-call rotations
Mentor junior engineers
Automate fleet operations including provisioning, monitoring, and remediation
Create and maintain operational documentation and runbooks
What you’ll need
5+ years designing, architecting, and scaling HPC or GPU compute clusters in production
Deep understanding of server hardware: GPUs, NICs, PCIe, memory, thermals, and power
Full-stack debugging across hardware, OS, drivers, and networking layers
Strong Linux administration with kernel drivers, RDMA tuning, and performance analysis
Experience with schedulers and orchestration such as Slurm and Kubernetes
Willingness to work in SF office 3-4 days per week and travel domestically
Original job description
sfcompute is building a company to de-risk the largest infrastructure build-out in history by creating a liquid market for GPU offtake. The HPC/ GPU Cluster Architect will be responsible for architecting and deploying GPU clusters, ensuring they run smoothly, and participating in on-call rotations while mentoring junior engineers.5+ years of experience…