GPU Distributed System Researcher
Core
Invent and implement novel programming models and runtime environments for high performance computing systems with heterogeneous processors and accelerators.
Role type
Senior IC GPU distributed system researcher
Builds
Novel programming models, runtime environments, and software stacks for AMD's future hardware accelerators
Domain
High Performance Computing (HPC), GPU systems, distributed systems
Deliverable
production ML models
Required skills
GPU programming, C/C++, Python, distributed systems, multi-threaded CPU applications, ML/HPC software, device driver development, MPI/OpenSHMEM, networking architectures, cluster-scale optimization, AMD ROCm stack, Git
Preferred skills
PhD in Computer/Electrical Engineering, patentable invention experience, networking simulation (NS-3, OMNeT++)
Technologies
ROCm, MPI, OpenSHMEM, NS-3, OMNeT++, Git
Responsibilities
Invent and design novel programming models and runtime environments; Implement innovative software solutions on prototype hardware; Drive research results into product roadmaps; Collaborate with product groups and vendors to improve ML/HPC ecosystem; Submit patentable inventions; Communicate research findings to academic and commercial audiences
Seniority
Senior, hands-on IC
