PhD Architecture Research Intern - GPU & Data-Center

NVIDIA AI; Seattle (WA); On-site; USD 52,000 - 129,000; Full time

Benefits offered by this job

Intern benefits

Job summary

NVIDIA invites a research-focused intern to shape the future of GPU and data-center computing. You will work across architecture, memory systems, and accelerators, building models, simulators, and prototypes with researchers and product teams.

You will apply computer architecture knowledge, parallel systems, and CUDA programming to explore hardware-software co-design and communicate insights through technical reports and presentations.

Qualifications

• Pursuing a PhD in a relevant discipline (CS/CE/EE/Physics/Math).
• Strong programming ability in C/C++ and scripting languages.
• CUDA programming experience and building system simulators.
• Background in computer architecture and parallel computer architectures.

Responsibilities

• Investigate new architecture concepts for future GPUs, memory systems, accelerators, and data-center-scale platforms.
• Study emerging workloads and identify architectural bottlenecks and opportunities.
• Explore hardware-software co-design across architecture, compilers, runtime systems and applications.
• Develop models, simulators, prototypes, and tools to evaluate architecture ideas.
• Collaborate with researchers, GPU architects, software teams and product groups.
• Communicate findings through presentations, reports, and potential publications.
• Help transfer research ideas and tools into NVIDIA product teams.

Skills

C/C++ programming

scripting languages

parallel computing

computer architecture

CUDA programming

Education

Pursuing PhD in CS/CE/EE/Physics/Math

Tools

CUDA toolkit

GPU profiling tools

system simulators

Job description

NVIDIA invites a research-focused intern to shape the future of GPU and data-center computing. You will work across architecture, memory systems, and accelerators, building models, simulators, and prototypes with researchers and product teams.

You will apply computer architecture knowledge, parallel systems, and CUDA programming to explore hardware-software co-design and communicate insights through technical reports and presentations.