【NVIDIA】招聘GPU Architect（C/C++)-上海、北京

NVHR

Hello 大家好,

这边是NVIDIA HR Tracy,目前我们GPU SW部门有如下职位招聘：

1.        在上海招聘GPU Architect（Graphics Shader方向）岗位，熟悉C++，熟悉计算机体系结构
2.        在上海招聘GPU Architect（compute/CUDA方向）熟悉C/C++，熟悉计算机体系结构
3.        在北京招聘  并行计算工程师( CUDA  Application)

有意向者 可以发送简历至tracyw@nvidia.com  QQ: 1751315121

1. GPU Architect (Shader/streaming) 方向
- PERFORMANCE /FUNCTIONAL MODELING, VALIDATIONAND ANALYSIS OF SHADER UNIT
NVIDIA's GPUs and SOCs are world leaders in performance and efficiency, and the SM (Streaming Multiprocessor or the Shader unit) is the core of its gigantic processing power. The SM architects define, design and implement the state of the art SM architecture powering next generation GPU. They proactively solve challenging technical problems and prudently tradeoff between performance, power and cost, to deliver the best result.
RESPONSIBILITIES:
- Investigate and propose architecture ideas based on quantitative study of existing and projected SM architecture.
- Develop performance and functional simulation models and simulation infrastructure.
- Develop performance and functional test plan and tests for new SM and architectural features.
- Test and debug on simulators, RTL and real silicon.
REQUIREMENTS:
- In depth knowledge of computer architecture, with good understanding of modern ISA and microprocessor implementation techniques.
- Good understanding of GPU concept and pipeline, interms of Graphics processing and parallel compute.
- Good mastery of C++ language and at least 3 years of C/C++ development experience.
- Experience of performance/functional modelling,profiling and analysis is a plus.
- Experience with trace-driven and execution-driven simulation model development is a plus.
- BSEE, BSCSE, or equivalent required. MS or PhD is a plus.
- Candidates with GPU shader unit design, microprocessor design and CPU/GPU performance analysis experience are

2.        Compute Architect (CUDA/HPC) –Shanghai
The NVIDIA GPU Computing Architecture group is seeking Regular candidates with backgrounds in computer architecture and computer science to join our effort to advance the state of parallel computing. Our team is designing the fastest and most efficient parallel architectures that power energy-efficient smart phones and the world's fastest super computers.

RESPONSIBILITIES:
•             Develop innovative HW, GPU and system architectures to extend the state of the art in performance, programmability, efficiency and reliability
Analyze and prototype key algorithms for new GPU architectures
•             Collaborate across the company to guide the direction of GPU computing, working with software, research and product teams
MINIMUM REQUIREMENTS:
•             Strong programming skills in C, C++, CUDA, Perl, Open MP, MPI or Python.
•             Strong background in computer architecture, compilers, parallel processing.
•             Experience in systems-level performance modeling, profiling, and analysis.
•             Experience in characterizing and modeling system-level performance, executing comparison studies, and documenting and publishing results.
•             Strong background in computer vision, machine learning is plus.

3.        Parallel Computing Software Engineer ( CUDA Application)-BJ
Description/Qualifications:         
To spearhead its breakthroughs in High Performance Computing, NVIDIA is looking for a passionate, world-class Software Engineer to work in its Developer Technology Team. In this role, you will open up new domains of application for GPUs by researching and developing GPU computing galgorithms, driving their adoption with key application developers and ensuring best possible performance of GPU computing applications on current and next-generation architectures.

Application domains include geosciences, life sciences, computer-aided engineering, computational fluid dynamics, computational chemistry, computational physics, computational finance, electronic design automation, data mining, medical imaging, and many more.
RESPONSIBILITIES:
- Work directly with key US developers on their scientific applications, educating them on the latest NVIDIA technologies.
- Closely collaborate with the architecture and software teams at NVIDIA to influence the design of next-generation architectures.
- Some travel to conferences and for on-site visits withed velopers will be required.

MINIMUM REQUIREMENTS:
- A good degree from a leading university in an engineering or computer science related discipline.
- Strong knowledge of programming languages such as Fortran and C/C++.
- Strong knowledge of programming techniques and algorithms.
- Strong mathematical fundamentals, including linear algebra and numerical methods.
- Experience with parallel programming, especially data-parallel using MPI, OpenMP, OpenACC or CUDA C/C++/Fortran.
- Good communication and organization skills, with alogical approach to problem solving, good time management and task prioritization skills.

With highly competitive salaries and a comprehensive benefits package, NVIDIA is widely considered to be one of the technology world’s most desirable employers. We have some of the most brilliant and talented people on the planet working for us and, due to unprecedented and ongoing growth, our world-class engineering teams are expanding fast. If you’re creative and autonomous Software Engineer with a genuine passion for parallel computing, we want to hear from you.


Tracy Wu
Staffing Team

1751315121