GPU Performance Engineer – AI Kernels
Core
Optimizing performance and power efficiency of machine learning kernels and AI workloads on AMD GPUs.
Role type
Senior IC GPU performance engineer (AI kernels)
Builds
High-performance AI software and GPU computing solutions
Domain
Semiconductor / High-performance computing / Artificial Intelligence
Deliverable
production ML models
Required skills
GPU architecture knowledge, C++ (C++17+), CUDA/HIP/OpenCL, GPU profiling and debugging, HPC, tile-based programming (Triton/CUTLASS), ML frameworks (PyTorch/MIOpen)
Preferred skills
Hardware ISA experience, CI/CD infrastructure improvement
Technologies
HIP, CUDA, OpenCL, Triton, CUTLASS, PyTorch, MIOpen, C++
Responsibilities
Optimize ML kernels for performance and power efficiency, research efficient alternatives for existing issues, improve development workflows and CI infrastructure, debug and resolve GPU software issues, stay informed on GPU architecture and ML algorithm trends
Seniority
Senior, hands-on IC
