GPU Software Specialist, Onboard Compute
Muon SpaceBase Salary
$156k - $186k/yr
Responsibilities
- Design and implement GPU compute kernels for onboard image processing, radio signal processing, and ML inference.
- Architect GPU pipelines that process live optical and radio sensor data and pass results to downlink or CPU decision-making subsystems.
- Port, adapt, and optimize customer code, models, and algorithms for real-time, power-constrained on-orbit execution.
- Profile and optimize GPU workloads, including occupancy, kernel launch configuration, and memory access patterns.
- Own unit, system-level, and hardware-in-the-loop verification and validation of GPU software.
- Define and operate the GPU build and CI/CD environment, including embedded-target cross-compilation toolchains and containerized builds.
- Collaborate with flight software, FPGA, payload, and hardware engineers on interfaces, data formats, and timing budgets.
- Translate mission and payload requirements into documented designs, trade studies, and interface specifications.
- Work with internal and external customers to guide algorithm and model adaptation for embedded GPU hardware.
Requirements
- Bachelor's or master's degree in Computer Science, Computer Engineering, Electrical Engineering, Applied Math, or a related technical field.
- At least three years of professional experience developing GPU-accelerated software.
- Strong proficiency in C/C++ or Rust and Python, including memory management, concurrency, and performance-oriented programming.
- Production experience shipping GPU-accelerated software with CUDA, OpenCL, HIP, or similar technologies.
- Deep knowledge of GPU architecture, SIMT execution, memory hierarchies, access patterns, occupancy, and kernel launch overhead.
- Experience developing and debugging on embedded Linux, including cross-compilation, device tree basics, and userspace/kernel driver interaction.
- Ability to write Linux userspace software integrating GPU compute through shared memory or similar mechanisms.
- Ability to work with customers to adapt algorithms and models for efficient execution on embedded GPU hardware.
- Strong written and verbal communication skills and ability to produce design documents, interface descriptions, and test reports.
- Ability and willingness to obtain and maintain a U.S. security clearance; active clearance is a plus.
- Experience with raw image or radio signal processing on GPUs is preferred.
- Experience deploying ML inference on GPUs, including quantization and edge model optimization, is preferred.
- Experience with DSP/RF workloads, Ethernet-based payload streams, RDMA, and zero-copy buffer techniques is preferred.
- Experience with embedded software CI/CD, containerization, and hardware-in-the-loop test automation is preferred.
- Space, aerospace, mission-critical software, customer-facing, or applied-engineering experience is preferred.
Benefits
- Hybrid work arrangement with three days per week on-site in the San Jose, California office.
- Medical, dental, and vision insurance.
- 401(k) retirement plan.
- Short- and long-term disability and life insurance.
- Three weeks of paid vacation for new employees.
- 12 paid holidays, unlimited sick time, and paid parental leave.
- Equity compensation is provided in addition to salary.
About Muon Space
Founded in 2021, Muon Space is an end-to-end space systems company that designs, builds, and operates mission-optimized satellite constellations to deliver critical data and enable real-time compute and decision-making in space. Its proprietary technology stack, Halo™, integrates advanced spacecraft platforms, robust payload integration and management, and a powerful software-defined orchestration layer to enable high-performance capabilities at unprecedented speed – from concept to orbit. With state-of-the-art production facilities in Silicon Valley and a growing track record of defense, civil, and commercial and national security customers, Muon Space is redefining how mission-critical intelligence is delivered from space. For more information on Muon Space, visit: https://www.muonspace.com/