Accelerated Compute Systems Performance Architect Intern - 2027
Core
Analyze and optimize performance of current and next-generation NVIDIA GPUs for high-performance computing, scientific computing, machine learning, and AI applications.
Role type
Intern, accelerated compute systems performance architect
Builds
Optimized GPU software-hardware co-design for datacenter and automotive computing
Domain
Hardware/Software architecture for accelerated computing (GPUs)
Deliverable
production ML models
Required skills
GPU programming (CUDA/OpenCL), C/C++, computer systems architecture, performance analysis and optimization, multi-node communication (MPI/OpenSHMEM/NVSHMEM)
Preferred skills
Python, patent writing, conference publications
Technologies
CUDA, OpenCL, MPI, OpenSHMEM, NVSHMEM, C, C++, Python
Responsibilities
Perform in-depth analysis and optimization to ensure best possible performance on NVIDIA GPUs; Understand and analyze the interplay of hardware and software architectures on core algorithms and programming models; Collaborate with hardware design, software engineering, product, and research teams; Dive into accelerated computing applications to facilitate software-hardware co-design; Write up and present work via white papers, conference publications, blog posts, or patent applications.
