GPU Kernel Development Engineer - Fellow
Core
Define and drive end-to-end software optimization strategy to achieve industry-leading performance for AMD's top-tier customers in GPU architecture, API design, and deployment.
Role type
Fellow, GPU kernel development and software optimization
Builds
High-performance ROCm core software for AI and HPC workloads
Domain
Semiconductor / AI / High-Performance Computing
Deliverable
production ML models
Required skills
GPU kernel optimization, machine learning software development, performance analysis, algorithm development, modern model architectures (Transformer, Attention, KV Cache), quantization, speculative decoding, FlashAttention
Preferred skills
GPU architecture knowledge, cross-functional initiative driving, experience with AI software stack
Technologies
ROCm, CUDA (implied via context), Transformer models
Responsibilities
Perform external benchmarking and research to improve GPU kernel performance; Analyze and optimize GPU kernels across the AI software stack; Partner with customers and hyperscalers to deliver tailored architectural wins; Collaborate with hardware, compiler, and framework teams to influence silicon features; Communicate with technical experts to improve ROCm applications and tools
