System and Scheduling Architect – GPU

NVIDIA NVIDIA · Semiconductors · Santa Clara, CA

NVIDIA is seeking a GPU system and process scheduling architect to design the next generation GPU system interface. This role involves understanding GPU sub-system interoperation, architecting features with cross-functional teams, creating performance models, and supporting post-silicon validation. The role requires experience in C/C++ and computer architecture, with knowledge of DL/ML being a plus.

What you'd actually do

  1. As a member of the architecture team, you will be responsible for understanding how the GPU sub-system modules interoperate and interact with overall system
  2. You will architect and plan features in concert with software, hardware, and verification teams across the globe to implement the next iteration of GPUs
  3. Create functional and performance models, often in C++, along with test plans to validate the features you design
  4. Actively support post-silicon validation activities

Skills

Required

  • BS or MS in Compute Architecture, Computer Science, or Electrical Engineering (or related degree) or equivalent experience
  • 2+ years of meaningful work experience
  • Work experience coding in C or C++
  • Course work or industry experience in Computer Architecture, Operation systems, and/or system architecture

Nice to have

  • Knowledge of memory isolation and security, firmware, process scheduling, OpenGL, DX, CUDA, DL/ML, PCIe, AXI, NVLink, and/or CXL
  • Knowledge of Perl or Python