Senior Staff Software Engineer, Linux Runtime & Drivers - San Francisco, Bay Area OR Pittsburgh, PA, United States
$220,000 - $270,000 USD yearly
Originally posted 6 October 2026 by the employer.
Contribute to the user-mode components that interface with the Fabric architecture and support the effcc compiler.
About the role
This role involves owning the user-mode components that interface with the Fabric architecture and supporting the effcc compiler.
What you'll do
- Design, implement, and maintain the user-mode runtime driver for the Fabric accelerator, including context management, program loading, and resource management.
- Define and implement user-mode interfaces between the compiler, Fabric ABI, and kernel mode driver.
- Design and implement underlying runtime library support for 3rd party ML framework backends like PyTorch, ONNX, or TensorFlow.
- Lead user-mode bring-up on first silicon, utilizing functional simulation, emulation, and FPGA environments.
- Establish runtime test coverage, including API unit tests, emulation-based regression suites, and hardware-in-the-loop validation.
What you'll need
- 7+ years of systems software experience, including hands-on Linux user-mode driver and language runtime library development for accelerators (GPUs, NPUs, DSPs, FPGAs).
- Experience building or contributing to a production accelerator runtime software stack, from driver API to user-facing runtime library.
- Solid understanding of accelerator memory models, unified memory systems, allocators, and zero-copy data paths.
- Experience designing public APIs for runtime libraries, focusing on ABI stability, versioning, error models, and thread safety.
- Experience integrating runtimes with language runtimes or ML frameworks (Python bindings, framework backends, C/C++ application interfaces).
- Hands-on experience with QEMU, emulation, or simulation environments during pre-silicon development.
- Expert C and C++ and strong Python skills; comfortable with AI-assisted development tools.
- Working knowledge of ARM/AArch64 Linux systems and the Linux driver model.
Nice to have
- Experience contributing to CUDA Runtime, ROCm, Tenstorrent runtime (TT-Metalium), oneAPI Level Zero, or OpenCL.
- Familiarity with LLVM or MLIR and compiler handoff to runtimes for accelerators.
- Knowledge of dataflow, spatial, or novel compute architectures.
- Experience with FPGA prototyping and hardware emulation platforms (Palladium, Zebu, Veloce).
- Experience on a first-generation silicon product where the software stack and hardware matured together.
This role has been open 9 days — well below the 60-day median for Systems Software Engineering roles.
Systems Software Engineering · Systems Software
|
Open roles in category
692
|
Median days open
60 d
|
Median salary
$213k
|
See the full market breakdown ▾Category comparison, and who else is hiring
| Metric | Efficient Computer | All employers we track in this specialty (692 roles · 71 employers) |
|---|---|---|
| Open roles in this specialty | 3 | 692 |
| Open roles in the wider Software, Firmware & Systems family | 4 | 6156 · 145 employers |
| Median days open | 9 d | 60 d (−51 d vs this employer) |
| Median salary (USD postings) | — | $213k |
Who's hiring in this category
- Qualcomm · 172 open roles · median 78 d
- NVIDIA · 135 open roles · median 48 d
- AMD · 63 open roles · median 50 d
- Mobileye · 27 open roles · median 71 d
- Broadcom · 21 open roles · median 52 d
- Arm Holdings · 20 open roles · median 113 d
How we counted: 692 open Systems Software Engineering (Systems Software) roles from 71 employers tracked in the SemiconductorJobs index, counted 16 Oct 2026. Specialty figures count only roles carrying this exact specialty label, so an employer's related work in neighbouring specialties is not included there — it is counted in the wider Software, Firmware & Systems family row. Figures refresh nightly.