GPU System Software Engineer: Performance & Profiling Tools

NVIDIA; Town of Texas (WI); On-site; USD 180,000 - 287,500; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA is seeking a system software engineer to build performance analysis capabilities for Nsight tools and the GPU Foundations Developer Tools team. You will develop foundational libraries, tests, and validation frameworks across CUDA and GPU assembly, collaborating with hardware architecture.

Strong C/C++ and Python skills, plus low-level assembly experience, are essential for this role. Join NVIDIA to work on scalable datacenter technologies, contribute to performance triage, and participate

Qualifications

• BS in EE/CS or equivalent experience; MS or PhD are also acceptable.
• 5+ years of experience or 2+ years with MS in relevant field.
• Strong programming ability in C, C++, and Python.
• Good understanding of low level assembly and hardware pipelines.
• Experience with performance analysis and optimization on hardware accelerators.
• Experience with simulators and emulators; collaboration in a team.

Responsibilities

• Design, develop, and maintain performance foundation libraries for Nsight tools.
• Develop GPU assembly tests and a test validation framework in CUDA and assembly.
• Utilize emulators to debug instruction events and verify timings.
• Document use cases and data workflows for architectural exploration.

Skills

C

C++

Python

Assembly

Problem solving

Team collaboration

Communication

Education

B.S. EE/CS (or equivalent experience)

MS

PhD

Tools

CUDA

Emulators

Job description

NVIDIA is seeking a system software engineer to build performance analysis capabilities for Nsight tools and the GPU Foundations Developer Tools team. You will develop foundational libraries, tests, and validation frameworks across CUDA and GPU assembly, collaborating with hardware architecture.

Strong C/C++ and Python skills, plus low-level assembly experience, are essential for this role. Join NVIDIA to work on scalable datacenter technologies, contribute to performance triage, and participate