What you’ll be doing:

• Work with multi-disciplinary teams to design, implement, and verify performance metrics and collection methodologies for NVIDIA GPUs/SoCs.
• Invent and improve our GPU/SoC profiling library with new features to allow developers to extract the best performance out of their applications.
• Understand and use the NVIDIA GPU/SoC performance monitoring system
• Get an opportunity to develop new tools for GPUs/SoCs running on several platforms.

What we need to see:

• B.S. EE/CS or equivalent with 4+ years of work experience or MS with 2+ years experience, or Ph.D.
• Strong programming ability in C, C++.
• Knowledge of computer architecture (preferably ARM CPUs/ SoCs or GPUs)
• Solid understanding of hardware pipeline concepts, with a willingness to work at a detailed implementation level
• Experience working in device drivers, embedded systems, or other low level library development

Ways to stand out from the crowd:

• Background including shipping device drivers or system software
• Knowledge of GPU API's such as: CUDA, OpenCL, OpenGL, Direct3D, Vulkan
• Prior experience authoring developer tools, particularly for GPUs or SoCs
• Experience in performance analysis, particularly of embedded system applications
• You have worked on computer algorithms and proven ability to choose the best possible algorithms to nail complex problems
• Knowledge or experience in compute architecture, operating systems, and compilers.