Lead HPC/GPU Cluster Architect | Automate & Scale

The San Francisco Compute Company; Boston (MA); Hybrid; USD 140,000 - 200,000; Full time

Benefits offered by this job

Generous equity grant

Visa Sponsorships

Retirement matching

Medical, dental & vision insurance

Time off & holidays

Parental leave

Daily lunch

Unlimited office book budget

Job summary

The San Francisco Compute Company is building the next generation of high-performance GPU compute infrastructure. You will design, deploy, and operate production HPC clusters, scaling to thousands of GPUs and ensuring reliability across hardware, OS, and networking layers.

You will drive automation, implement infra-as-code, and mentor junior engineers while collaborating with a small, ambitious team in a hybrid SF office model 3–4 days per week and occasional travel.

Qualifications

• 5+ years of hands-on experience designing, architecting and scaling HPC or GPU compute clusters in production.
• Strong understanding of server hardware fundamentals (GPUs, NICs, PCIe, memory, thermals, power).
• Ability to debug performance and reliability across hardware, OS, drivers, networking.
• Interest in automating fleet operations and infrastructure-as-code.
• Ability to document runbooks and operational procedures.

Responsibilities

• Participate in on-call rotation and keep GPU clusters running smoothly.
• Deploy new environments and scale deployments across regions.
• Diagnose and fix issues across hardware, OS, drivers, and networking layers.
• Develop automation to reduce manual toil and improve reliability.
• Mentor junior engineers and contribute to team culture.

Job description

The San Francisco Compute Company is building the next generation of high-performance GPU compute infrastructure. You will design, deploy, and operate production HPC clusters, scaling to thousands of GPUs and ensuring reliability across hardware, OS, and networking layers.

You will drive automation, implement infra-as-code, and mentor junior engineers while collaborating with a small, ambitious team in a hybrid SF office model 3–4 days per week and occasional travel.