Senior GPU Software Engineer - AI & HPC (ROCm)
Core
Design, develop, and optimize high-performance GPU kernels and features for machine learning and high-performance computing workloads.
Role type
Senior GPU Software Architect (AI & HPC)
Builds
ROCm software stack, GPU kernels, and high-performance compute solutions
Domain
Semiconductor / AI / High-Performance Computing
Deliverable
production ML models
Required skills
GPU kernel development, performance optimization, HIP, CUDA, machine learning frameworks, HPC libraries, parallel computing architectures
Preferred skills
DSP software, AI accelerators, algorithm development, profiling and debugging, technical documentation
Technologies
HIP, CUDA, ROCm, Convolution, GEMM, Attention
Responsibilities
Design and develop high-performance GPU kernels for ML and Convolution workloads; Collaborate with GPU architects to explore emerging hardware features; Analyse, profile, and optimize GPU kernel performance; Create technical documentation and share best practices; Partner with engineers and customers to enhance ROCm applications and tools
Seniority
Senior, hands-on IC
