HPC/ GPU Architect

Acceler8 Talent; San Francisco (CA); Hybrid; USD 180,000 - 240,000; Full time

Job summary

Acceler8 Talent is seeking a Staff / Senior Staff HPC & GPU Cluster Architect to design and own production GPU clusters. You will translate customer workloads into hardware sizing and a complete bill of materials, working closely with the CTO.

You will validate network topology, mentor engineers, and bring experience with large GPU deployments (1,000+ GPUs a strong plus) while traveling as needed in the US.

Qualifications

• Designed and delivered an interconnected GPU cluster in production.
• Strong knowledge of current NVIDIA platforms and the AI stack.
• Mentoring engineers on GPU hardware architectures.
• Experience with 1,000+ GPUs is a strong plus.

Responsibilities

• Design and own production GPU clusters, translating workloads into hardware sizing and BOM.
• Validate network topology and ensure proper power and cooling.
• Collaborate with CTO and engineers on architecture decisions.
• Travel as needed to support installations and escalations.

Skills

GPU cluster design

NVIDIA platforms

Network validation

Mentoring

Job description

Hiring: Staff / Senior Staff HPC & GPU Cluster Architect

I’m partnering with a GPU infrastructure startup hiring a founding hardware architect to work directly with the CTO.

You’ll own the design of production GPU clusters, translating customer workloads into hardware sizing and a complete bill of materials.

What we’re looking for:

• Personally designed and delivered an interconnected GPU cluster in production.
• Made the GPU, server and fabric decisions, with practical understanding of power and cooling.
• Strong knowledge of current NVIDIA platforms and the AI stack.
• Confidence validating network topology and mentoring engineers.

Experience with 1,000+ GPUs is a strong plus.

US-based, with travel required and occasional support for complex incident escalations.