Research Scientist, System Software & I/O for GPUs

Experience Champaign Urbana; Champaign (IL); On-site; USD 192,000 - 356,000; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA is seeking a Research Scientist focused on System Software and I/O Architecture to advance scalable storage access by GPU threads. You will co-optimize architecture, runtime systems, OS and compilers for high throughput and energy efficiency.

The role requires a Ph.D. and 5+ years of related research, with strong CUDA/C/C++/Python skills, and experience in profiling and large-scale clusters. Equity and benefits are offered, with applications open through January 13, 2026.

Qualifications

• Ph.D. in CE/CS/EE or equivalent experience.
• 5+ years of research in computer architecture, OS, system administration, compilers, and/or HPC.
• Publication, patent or collaboration history is a big plus.
• Expertise in a specific area and ability to become go-to resource.
• Experience with experimental architecture research and software infra.
• Open-source software release track record.
• Proficiency in C, C++, CUDA, Python and scripting; MPI/NACL a plus.
• Strong interpersonal and presentation skills.
• Experience with workloads like recommender systems, graph analytics, data frames.

Responsibilities

• Develop novel architectures and system software for scalable multi-GPU platforms.
• Analyze OS, CPU/GPU interplay and efficient algorithm designs.
• Collaborate with software research, hardware engineering and product teams.
• Publish original research and present at conferences and events.

Skills

CUDA

C/C++

Python

Scripting languages

MPI

NACL

Interpersonal skills

Education

PhD in CE/CS/EE

Tools

Profiling tools

Job description

NVIDIA is seeking a Research Scientist focused on System Software and I/O Architecture to advance scalable storage access by GPU threads. You will co-optimize architecture, runtime systems, OS and compilers for high throughput and energy efficiency.

The role requires a Ph.D. and 5+ years of related research, with strong CUDA/C/C++/Python skills, and experience in profiling and large-scale clusters. Equity and benefits are offered, with applications open through January 13, 2026.