Senior HPC Systems Architect — AI Cloud Compute

Showcify; United States; On-site; USD 180,000 - 240,000; Full time

Benefits offered by this job

Health, dental, and vision coverage

Wellness and commuter stipends

401k with 2% company match

Flexible paid time off

Job summary

Lambda is seeking a senior compute architect to lead the design of scalable AI/ML compute platforms in a hybrid office model. The role focuses on optimizing GPU-based infrastructure, collaborating with product and engineering teams to meet workload requirements, and guiding platform validation.

A strong background in architecture and performance is essential. The position involves working across hardware and software boundaries, mentoring engineers, and shaping roadmaps for hardware and rack

Qualifications

• 7+ years in architecting large GPU HPC/cloud compute platforms.
• Deep knowledge of CPU/GPU architectures and memory hierarchies.
• Experience with high-bandwidth fabrics (NVLink/InfiniBand/RoCE).

Responsibilities

• Architect and define scalable compute platforms for AI/ML workloads.
• Develop compute system standards for performance and maintainability.
• Evaluate CPU/GPU/accelerator technologies and design tradeoffs.
• Collaborate to map workload requirements to compute capabilities.
• Translate ambiguous needs into measurable platform requirements.
• Define roadmaps and reference designs for hardware and racks.
• Act as technical lead during new platform introductions.
• Mentor engineers on compute performance and sizing.

Skills

GPU HPC architecture

CPU/GPU architectures

High-bandwidth interconnects

Performance tuning

Hardware-software collaboration

Architectural tradeoffs

Analytical communication

Ownership / self-starter

Tools

Slurm

Kubernetes

GPU virtualization

Job description

Lambda is seeking a senior compute architect to lead the design of scalable AI/ML compute platforms in a hybrid office model. The role focuses on optimizing GPU-based infrastructure, collaborating with product and engineering teams to meet workload requirements, and guiding platform validation.

A strong background in architecture and performance is essential. The position involves working across hardware and software boundaries, mentoring engineers, and shaping roadmaps for hardware and rack