AI/ML Compiler Developer (NPU Acceleration)
Core
Develop AI/ML specific C/C++ kernels and dataflow schedules for AMD Ryzen Processors with XDNA NPUs to map LLMs and Stable Diffusion networks.
Role type
Staff Software Engineer (AI/ML Compiler & Kernel Developer)
Builds
Optimized C++ kernel libraries and dataflow schedules for vector processors (NPU/GPU)
Domain
Semiconductor / AI Hardware Acceleration
Deliverable
production ML models
Required skills
C++, Python, SIMD programming, VLIW architecture, parallel computing, performance profiling, unit testing, low-level hardware optimization
Preferred skills
TensorFlow, PyTorch, silicon bring-up, emulation platforms, cache hierarchy knowledge
Technologies
C++, Python, TensorFlow, PyTorch, Git, XDNA NPU, VLIW, SIMD
Responsibilities
Design and implement optimized C++ kernel libraries for NPU/GPU; Develop vectorized code leveraging SIMD and ILP; Profile and tune kernel performance; Write unit and integration tests; Collaborate with hardware and research teams; Document design specs and performance improvements
Seniority
Staff, hands-on IC
