What you’ll be doing:
Design, develop and maintain our GPU performance foundation library for Nsight tools.
Design, develop performance triages for upcoming and latest LLM chips.
Collaborate closely with the Hardware Architecture team to co-design and optimize software and hardware interfaces.
Define, invent, and improve our GPU profiling library with new features to allow NVIDIA's customers to extract the best performance out of their code base.
Design and implement test plans to validate the performance and functionality of the software libraries.
Stay up-to-date with the latest advancements in LLM inference, hardware acceleration, and software optimization techniques.
What we need to see:
B.S. EE/CS or equivalent experience with 4+ years of experience or MS with 2+ years' experience, or Ph.D.
Strong programming ability in C, C++, and scripting languages.
Solid understanding of hardware pipeline concepts, with a willingness to work at a detailed implementation level.
Experience with performance analysis and optimization of software on hardware accelerators.
Knowledge of hardware-software co-design principles and practices.
Excellent problem-solving skills and the ability to work collaboratively in a team environment.
Strong communication skills, both written and verbal.
Ways to stand out from the crowd:
Proven knowledge of compute (CUDA/OpenCL) and/or graphics (DirectX, OpenGL, Vulkan).
Prior experience authoring developer tools, particularly for GPUs, games, or pro visualization.
You will also be eligible for equity and .
משרות נוספות שיכולות לעניין אותך