GPU Development Tools Engineer
岗位可以 base 北京或者上海
可以发送简历到邮箱-xiaozhao@nvidia.com
加微信了解详细情况:18867144803
As a member of the software development team, you will engineer and improve the core infrastructure for execution, automation, and debugging the development of large-scale, general-purpose graphics and computing chips. This infrastructure enables our driver stack, applications, tests, and studies to run unchanged on all functional, diagnostic, and performance models.
[ What you’ll be doing:]
This role will require you to play a critical part in every stage of development of a GPU!
Improve the daily workflows of the world’s top chip modelers and designers to help produce the next greatest generation of GPUs.
Empower GPU architects to understand application performance today and model competition-destroying performance for tomorrow.
Coordinate with architecture and software teams to enable functional and performance testing for the next architecture.
[ What we need to see:]
Bachelor's or higher degree in Computer Science or Computer Engineering or equivalent experience
5+ yrs of experience
Strong C++, or Java and JavaScript programming capability
Aptitude to work across the GPU, driver, and application stacks
Some familiarity with a scripting language, such as Python
Excellent interpersonal skills
Flexibility for working in an evolving environment with different frameworks and requirements
[ Ways to stand out from the crowd:]
Know-how working on operating system kernels or writing device drivers with strong systems-level debugging skills
A knowledge of GPU APIs such as DirectX, CUDA, Vulkan or OpenGL
Experience with chip and/or system simulation
Experience in performance analysis of sophisticated systems
Deep understanding of systems architecture: CPU, GPU, memory, display, buses, kernel internals would be helpful
这是一个专为移动设备优化的页面(即为了让你能够在 Google 搜索结果里秒开这个页面),如果你希望参与 V2EX 社区的讨论,你可以继续到 V2EX 上打开本讨论主题的完整版本。
V2EX 是创意工作者们的社区,是一个分享自己正在做的有趣事物、交流想法,可以遇见新朋友甚至新机会的地方。
V2EX is a community of developers, designers and creative people.