Sr. Staff AI System Software Engineer

Location: Hsinchu, Taiwan. The role requires satisfactory proof of the right to work in Taiwan and authorization to access export-controlled technology.

Company

hirify.global develops high-performance RISC-V compute platforms for artificial intelligence, machine learning, automotive, data center, mobile, and consumer applications.

What you will do

• Design and develop Linux system software for RISC-V AI accelerator platforms, including kernel components, device drivers, DMA engines, and memory-management subsystems.
• Develop low-level device runtimes and heterogeneous memory software for efficient communication and memory sharing between RISC-V host processors and AI accelerators.
• Optimize PCIe, CXL, IOMMU, ATS, PRI, PASID, and related host-to-accelerator communication mechanisms.
• Integrate accelerator runtimes with PyTorch, TVM, MLIR, and related machine-learning software stacks.
• Analyze system performance and create benchmarks, profiling tools, tracing infrastructure, and regression tests for AI workloads.
• Debug issues across user space, the Linux kernel, firmware, FPGA prototypes, and accelerator hardware while collaborating with architecture, hardware, firmware, compiler, and software teams.

Requirements

• 8+ years of experience developing architecture-level code, operating-system components, or device drivers in C or C++ for multiprocessor and multithreaded systems such as Linux or BSD.
• Strong knowledge of computer architecture, virtual memory, DMA, cache coherency, address translation, interrupts, and synchronization.
• Experience with device drivers, memory management, DMA, interrupts, virtualization, IOMMUs, and debugging complex multicore or heterogeneous systems.
• Experience with GDB, JTAG, OpenOCD, perf, ftrace, or equivalent debugging and profiling tools.
• Experience with Git, Makefiles, the GNU toolchain, and shell scripting.
• B.Sc. or M.Sc. in Computer Science, Computer Engineering, Electrical Engineering, or a related discipline, or equivalent practical experience.

Nice to have

• Experience developing firmware or software for GPUs, NPUs, TPUs, or other AI accelerators.
• Experience with RISC-V architecture, paging, privilege levels, interrupts, exceptions, SBI, or platform firmware.
• Experience with heterogeneous memory systems such as CUDA Unified Memory, AMD ROCm, Linux HMM, or OpenCL Shared Virtual Memory.
• Experience integrating custom accelerator runtimes with PyTorch, TVM, MLIR, or other machine-learning frameworks.
• Experience with hardware performance counters and system-level profiling tools.

Culture & Benefits

• Collaborate with architecture, hardware, firmware, compiler, and software engineering teams.
• Contribute to an open and programmable AI computing platform based on the RISC-V instruction set architecture.
• Employment is contingent on successful background and reference checks.
• Export-controlled technology access requires applicable authorization or successful approval of any necessary export license.
• Inclusive workplace committed to diversity and equal opportunity.