Senior HPC Systems Architect: Liquid-Cooled GPU Cloud

Socket.dev; San Jose (CA); Hybrid; USD 180,000 - 260,000; Full time

Job summary

Lambda, The Superintelligence Cloud, is seeking a Senior HPC Systems Architect to design and lead the development of large-scale GPU clusters and HPC infrastructure. This role requires presence in the San Francisco or San Jose office 4 days per week, with Tuesday as the designated work-from-home day.

You will define system requirements, build testing frameworks, evaluate new technologies, and mentor engineering teams to deliver scalable, production-ready solutions for AI workloads.

Qualifications

• 8+ years of experience designing and architecting large-scale HPC and distributed computing systems.
• Expert-level knowledge of HPC hardware including GPU clusters, compute nodes, InfiniBand and Ethernet, and distributed storage.
• Hands-on experience with direct-to-chip liquid cooling systems.
• Proven expertise in creating robust performance benchmarks, capacity planning, and system validation.
• Exceptional skills in system architecture, design documentation, and technical specifications.
• Ability to work collaboratively across teams, ensuring alignment of technical solutions with business objectives.

Responsibilities

• Design and architect advanced HPC systems optimized for large-scale computational workloads and AI applications.
• Collaborate with internal teams and stakeholders to define system requirements and performance goals.
• Develop comprehensive testing frameworks to rigorously assess system performance, scalability, and reliability.
• Evaluate emerging technologies and architectural approaches to continuously enhance infrastructure capabilities.
• Create detailed architectural plans, documentation, and blueprints to guide implementation teams.
• Provide technical leadership and mentoring to engineering teams, fostering best practices in HPC architecture.

Skills

HPC architecture

GPU clusters

High-speed networking

System benchmarking

Documentation

Cross-team collaboration

Problem solving

Tools

Ansible

Terraform

Kubernetes

InfiniBand

Job description

Lambda, The Superintelligence Cloud, is seeking a Senior HPC Systems Architect to design and lead the development of large-scale GPU clusters and HPC infrastructure. This role requires presence in the San Francisco or San Jose office 4 days per week, with Tuesday as the designated work-from-home day.

You will define system requirements, build testing frameworks, evaluate new technologies, and mentor engineering teams to deliver scalable, production-ready solutions for AI workloads.