Senior HPC Network Architect for GPU Clusters

SF Compute; San Francisco (CA); Ibrido; USD 180.000 - 260.000; Tempo pieno

Vantaggi offerti da questo lavoro

Generous equity grant

Visa Sponsorships

Retirement matching (401k)

Medical, dental & vision insurance

Unlimited paid time off

Parental leave

Daily lunch

Unlimited office book budget

Descrizione del lavoro

San Francisco Compute seeks a senior Network Architect to lead the design and operation of world-spanning GPU clusters. You will architect HPC networks, drive automation, and mentor engineers while contributing to a fast-growing, high-performance compute platform.

You will work with a small, ambitious team and participate in on-call rotations, shaping culture and collaborating with customers to deliver scalable, reliable compute solutions from our San Francisco base.

Competenze

• 10+ years of hands-on management or architecture with network for at least one GPU cluster
• Deep understanding of Ethernet (RoCEv2) and/or InfiniBand networks in CLOS/fat-tree topologies
• Experience building HPC network architectures (eBGP, fat-tree, VXLAN, MCLAG)
• Interest in implementing zero-touch provisioning for large multi-layer networks and automation
• Appreciates good documentation and clear standards
• Ability and willingness to mentor junior engineers
• Willing to come to the San Francisco office 3-4 days per week

Mansioni

• Architect and deploy new GPU clusters globally and ensure reliable operation
• Participate in on-call rotations and rapidly fix issues
• Lean into automation to enable scalable deployments
• Shape team culture and mentor junior engineers
• Engage with customers and stakeholders to tailor network solutions

Conoscenze

10+ years experience

Ethernet RoCEv2/InfiniBand

HPC network architectures

Automation & zero-touch provisioning

Documentation

Mentoring engineers

SF office 3-4 days/week

Strumenti

Linux systems administration

Linux virtualization (KVM/QEMU/libvirt)

Containers & Kubernetes operators

Descrizione del lavoro

San Francisco Compute seeks a senior Network Architect to lead the design and operation of world-spanning GPU clusters. You will architect HPC networks, drive automation, and mentor engineers while contributing to a fast-growing, high-performance compute platform.

You will work with a small, ambitious team and participate in on-call rotations, shaping culture and collaborating with customers to deliver scalable, reliable compute solutions from our San Francisco base.