Senior Software Engineer, Cloud Image/ML Compute Architecture
About this role
Muon Space needs to scale its high-performance compute infrastructure to process a growing volume of scientific imaging and machine learning workloads across ground and flight systems. The high-performance compute team produces critical image data products, and this engineer will design and deploy the serving platform that runs these models in production.
Day-to-day responsibilities focus on architecture and performance optimization, establishing the patterns needed to transition research models into low-latency production services. Candidates who combine expert C++ abilities with hands-on PyTorch and distributed systems experience tend to thrive in this hybrid San Jose environment.
What you will do
- Architect, develop, and deploy the cloud image processing and machine learning serving stack for ground data products.
- Design distributed and low-latency image processing systems that scale effectively in the cloud environment.
- Define patterns and interfaces allowing data pipelines to interact with performant native and accelerated code.
- Set up production-grade serving paths for image data pipelines while extending architectures for future onboard missions.
- Deliver high-performance reference algorithm implementations and conduct thorough benchmarking.
- Mentor team members on modern C++, machine learning runtimes, and distributed systems design.
- Manage verification and validation processes including unit, integration, and performance regression tests.
What you need
- Hold a Bachelor's or Master's degree in Computer Science, Computer Engineering, Electrical Engineering, Applied Math, or a related field.
- Bring 6 or more years of professional software engineering experience featuring technical leadership in distributed computation or data-intensive systems.
- Demonstrate expert-level C++ skills with a deep understanding of memory management, concurrency, and performance-oriented programming.
- Possess strong PyTorch experience alongside production background running models in C++ cloud runtimes using LibTorch or equivalent.
- Show demonstrated experience in architecting distributed systems for low-latency image processing, machine learning, or computer vision.
- Maintain strong Python capabilities, particularly when interacting with native optimized implementations.
- Exhibit strong written and verbal communication skills to lead cross-functional technical discussions.
- Hold the ability and willingness to obtain and maintain a U.S. security clearance.
Nice to have
- Experience with accelerators beyond GPUs such as TPUs and reasoning about accelerator tradeoffs
- Model optimization for inference including quantization, graph compilation, and kernel fusion
- Production ownership for latency-critical production systems
- GPU acceleration using CUDA, HIP, or similar frameworks
- CI/CD for containerized workloads with performance regression testing
- Direct exposure to space, aerospace, or mission-critical software environments
Benefits
- Competitive equity grant
- Medical, dental, and vision insurance
- 401k retirement plan
- Short & long term disability and life insurance
- Three weeks paid vacation for new employees, 12 paid holidays, and unlimited sick time
- Paid parental leave
Skills and tech stack
Questions candidates ask
What is the work location and remote policy for this role?
This position is hybrid and requires working on-site in the San Jose, California office three days per week.
What is the salary range for this position?
The salary range for this role is $232,000 to $261,000, plus a competitive equity grant and comprehensive benefits package.
What tech stack and programming languages are used?
The tech stack relies on expert-level C++, strong Python skills, and experience with PyTorch and LibTorch for running models in cloud runtimes.
Are there specific citizenship or security clearance requirements?
Applicants must have the ability and willingness to obtain and maintain a U.S. security clearance. The role also requires compliance with U.S. Government export regulations.