Senior HPC Systems Engineer — Space GPU Fabric

Blue Origin; Hutto (TX); On-site; USD 134,000 - 203,000; Full time

Benefits offered by this job

Medical, dental, vision

401(k) with company match

Paid time off

Job summary

Blue Origin is seeking a hardware-leaning systems engineer to architect and build the nervous system of a large-scale orbital data center. You will own foundational hardware decisions for the high-speed interconnect fabric and manage its performance from the physical layer to the software stack.

You will solve first-principles challenges like fabric resiliency in radiation environments and work closely with software teams to ensure model-parallel AI workloads run efficiently on a multi-tenant

Qualifications

• B.S. in CS, EE, or related field.
• 8+ years in HPC networking, datacenter fabric design, or GPU cluster engineering.
• Experience with high-speed GPU interconnects and large-scale accelerator clusters.
• Hands-on experience with large-scale network topology and fabric management tools.
• Proficiency with collective communication libraries.
• US citizen, national, permanent resident or eligible to work in the U.S.

Responsibilities

• Architect hierarchical GPU interconnect topology for a large-scale accelerator cluster.
• Own configuration, bring-up, and operation of fabric management software.
• Lead hardware qualification of fabric switches for space environment; mitigate radiation effects.
• Own performance budgets for fabric bandwidth/latency for model-parallel inference.
• Collaborate with software teams to co-design topology with model parallelism.
• Interface with platform software teams to manage fabric bring-up and integration.

Skills

HPC networking

Datacenter fabric design

GPU cluster engineering

Network topology at scale

Collective communication libraries

Education

Bachelor of Science in Computer Science, Electrical Engineering, or a related field

Master's or Ph.D. in a relevant discipline

Tools

InfiniBand fabrics

Fabric management tools

NCCL / RCCL

Job description

Blue Origin is seeking a hardware-leaning systems engineer to architect and build the nervous system of a large-scale orbital data center. You will own foundational hardware decisions for the high-speed interconnect fabric and manage its performance from the physical layer to the software stack.

You will solve first-principles challenges like fabric resiliency in radiation environments and work closely with software teams to ensure model-parallel AI workloads run efficiently on a multi-tenant