System Software Engineer, Profiler - GPU

at Nvidia

USD 152,000-287,500 per year

MIDDLE SENIOR

✅ On-site

Tech Stack
Tag name is followed by "@" symbol and proficiency level value. About proficiency levels:

1-2 — basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;

3-6 — daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;

7-9 — you are an expert, you can teach others, you know all the pitfalls and tricks;

10 — exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.

CUDA @ 3 Communication @ 6 Deep Learning @ 3 GPU @ 3 HPC JAX @ 3 OpenCL @ 3 OpenGL @ 3 Performance Analysis @ 3 PyTorch @ 3 Python @ 6

Details

At NVIDIA, we build groundbreaking products for the automotive, virtual reality, gaming, deep learning, and high-performance computing sectors. As a member of the GPU Foundations Developer Tools team, you will help developers debug, profile, and analyze the performance of their systems and applications using low-level libraries. You will develop performance analysis capabilities for NVIDIA hardware, Nsight tools, and foundation libraries supporting next-generation accelerated computing at datacenter scale.

As a system software engineer in the Developer Tools group, you will develop software that enables GPU application developers to build compute applications deployed in the world’s largest distributed environments. The team is seeking a talented software engineer to contribute to performance-triage development and the co-design of foundational software libraries for developer tools in collaboration with the Hardware Architecture team.

Responsibilities

Design, develop, and maintain GPU performance foundation libraries for Nsight tools, focusing on high-fidelity hardware events and counters.
Develop and implement GPU assembly tests.
Build and maintain a test validation framework written primarily in CUDA and GPU Assembly.
Use emulators to debug and verify instruction events.
Document tool use cases and data-processing workflows to facilitate architectural explorations.

Requirements

Bachelor’s degree in electrical engineering or computer science, or equivalent experience, with 5+ years of experience; or a master’s degree with 2+ years of experience; or a Ph.D.
Strong programming ability in C, C++, and scripting languages such as Python.
Good understanding of, or prior experience with, low-level assembly code.
Solid understanding of hardware pipelines and execution-unit instruction-pipeline concepts, with a willingness to work at a detailed implementation level.
Knowledge of hardware-software co-design principles and practices.
Experience with performance analysis and software optimization on hardware accelerators.
Experience developing on simulators and emulators.
Excellent problem-solving skills and the ability to work collaboratively in a team environment.
Strong written and verbal communication skills.

Preferred Qualifications

Knowledge of compute technologies such as CUDA or OpenCL.
Knowledge of modern graphics technologies such as DirectX 12, OpenGL, Vulkan, or Metal.
Knowledge of deep-learning frameworks such as PyTorch or JAX.
Prior experience authoring developer tools, particularly for GPUs, games, professional visualization, or compute workloads.
Knowledge of GPU application performance analysis.
Experience in driver development.

Compensation and Benefits

Base salary range: $152,000–$241,500 for Level 3 and $184,000–$287,500 for Level 4, depending on location, experience, and compensation for similar positions.
Eligible for equity and benefits.
Applications will be accepted at least until July 28, 2026.
NVIDIA is an equal opportunity employer committed to an inclusive work environment.