Onboard GPU HPC Engineer for Space Missions

Muon Space; San Jose (CA); Hybrid; USD 156,000 - 186,000; Full time

Job summary

Muon Space is seeking a GPU Software Specialist to design and develop GPU-accelerated software that runs onboard satellites, powering mission-critical workloads like Earth-imaging sensor processing and onboard AI/ML inference.

You will work across the full development lifecycle, from feasibility to deployment, collaborating with FPGA, flight software, payload, and system hardware teams. This hybrid role requires on-site work in San Jose three days a week.

Qualifications

• Bachelor's or Master’s degree in CS/CE/EE or related field; 3+ years GPU software experience.
• Strong C/C++ or Rust and Python with memory management and concurrency.
• Production experience shipping GPU software using CUDA, OpenCL, or HIP.
• Deep knowledge of GPU architecture including memory hierarchies and occupancy.
• Experience on embedded Linux (e.g., Ubuntu on Jetson/IGX) with cross-compilation and drivers.
• Ability to translate customer algos/models to onboard GPU workloads; good communication.

Responsibilities

• Design and implement GPU compute kernels (e.g., CUDA) for onboard image and ML workloads.
• Architect end-to-end GPU pipelines ingesting live sensor data for on-orbit processing.
• Port and optimize code for embedded GPUs under real-time and power limits.
• Profile and optimize GPU workloads to meet deadlines within power budgets.
• Verify and validate GPU software with unit and system-level tests, including HITL tests.

Skills

C/C++

Rust

Python

GPU programming

Performance optimization

Linux userspace

Education

Bachelor's or Master's in CS/CE/EE or related

Tools

CUDA

OpenCL

HIP

Embedded Linux

Job description

Muon Space is seeking a GPU Software Specialist to design and develop GPU-accelerated software that runs onboard satellites, powering mission-critical workloads like Earth-imaging sensor processing and onboard AI/ML inference.

You will work across the full development lifecycle, from feasibility to deployment, collaborating with FPGA, flight software, payload, and system hardware teams. This hybrid role requires on-site work in San Jose three days a week.