Space GPU RAS Validation Lead

Cowboy Space Corporation; Seattle (WA); On-site; USD 150,000 - 225,000; Full time

Benefits offered by this job

Equity

Medical insurance

Dental insurance

Vision insurance

401(k)

Paid time off

Paid holidays

Parental leave

Relocation assistance

Lunch in office

Job summary

Cowboy Space Corp. is seeking a seasoned hardware validator to own the end-to-end RAS validation for GPU server systems deployed in Low Earth Orbit. You will partner with silicon vendors and ODMs to analyze failures, map fault paths, and verify detection and recovery across DDR, HBM, CPU, and PCIe.

Located in San Carlos or Seattle, this on-site role offers equity, comprehensive benefits, relocation support, and a fast-paced opportunity to shape fault-tolerance in space-grade compute.

Qualifications

• 5+ years in hardware validation or silicon validation on server-class compute systems.
• Deep knowledge of CPU/GPU architecture and memory subsystems.
• Hands-on fault injection across hardware, firmware and software levels.
• Experience with silicon vendors or ODMs on failure analysis.

Responsibilities

• Lead RAS validation strategy and execution for GPU server platforms, including fault injection and recovery verification.
• Collaborate with GPU system designers to analyze failures and align fault handling requirements for DDR, HBM, CPU, and GPU subsystems.
• Characterize fault propagation paths and ensure errors are correctly classified and logged.
• Validate BMC and out-of-band visibility via IPMI, Redfish, and MCTP/PLDM.
• Debug complex failure modes spanning GPU/CPU, memory, PCIe/NVLink, and firmware.

Skills

RAS validation

CPU/GPU architecture

Fault injection

Python scripting

System-level validation

Tools

BMC/IPMI

Redfish

MCTP/PLDM

Job description

Cowboy Space Corp. is seeking a seasoned hardware validator to own the end-to-end RAS validation for GPU server systems deployed in Low Earth Orbit. You will partner with silicon vendors and ODMs to analyze failures, map fault paths, and verify detection and recovery across DDR, HBM, CPU, and PCIe.

Located in San Carlos or Seattle, this on-site role offers equity, comprehensive benefits, relocation support, and a fast-paced opportunity to shape fault-tolerance in space-grade compute.