BSP & Fabric Platform Engineer for Space Compute

Blue Origin; Denver (CO); On-site; USD 182,000 - 254,000; Full time

Benefits offered by this job

Medical, dental, vision

Stock options

401(k) with company match

Paid time off

Job summary

Blue Origin is seeking a highly skilled platform software engineer to own the BSP and fabric bring-up for accelerator compute payloads. Responsibilities include bootloader, Linux kernel bring-up, and device drivers for a space-grade compute stack.

The role focuses on robust health monitoring, fault-tolerant recovery, and defining clear API boundaries with higher-level orchestrators and inference runtimes. This is a high-impact, hands-on position in a remote-ready space environment.

Qualifications

• BS in Computer Science, Computer Engineering, or related field.
• 6+ years in embedded BSP or low-level platform software.
• Hands-on with accelerator hardware software bring-up and validation.
• Strong C/C++ and Python for embedded systems; CUDA/ROCm familiarity.

Responsibilities

• Develop and own the BSP for accelerator compute modules including bootloader and kernel bring-up.
• Lead software bring-up and validation of high-speed compute fabric and GPU communications.
• Implement in-flight health monitoring: temperatures, ECC errors, power, interconnects.
• Design automated recovery for radiation-induced soft errors for resilience.
• Define API boundaries between platform software and higher-level services.

Skills

Embedded C/C++

Python

Linux kernel bring-up

GPU programming (CUDA/ROCm)

Driver/firmware debugging

Education

Bachelor's in CS/CE or related field

Master's/PhD preferred

Tools

Fabric management software

Pre-silicon emulation platforms

FPGAs

Job description

Blue Origin is seeking a highly skilled platform software engineer to own the BSP and fabric bring-up for accelerator compute payloads. Responsibilities include bootloader, Linux kernel bring-up, and device drivers for a space-grade compute stack.

The role focuses on robust health monitoring, fault-tolerant recovery, and defining clear API boundaries with higher-level orchestrators and inference runtimes. This is a high-impact, hands-on position in a remote-ready space environment.