Senior Deep Learning Hardware Modeling Architect - Lpu

NVIDIA NVIDIA · Semiconductors · CA +2 · Remote

NVIDIA is seeking a Senior Deep Learning Hardware Modeling Architect to optimize AI inference speed and efficiency. The role involves driving architectural specifications, developing written specifications for component-level and system-level designs, and embodying these specifications in an executable model. The candidate will ensure high performance using C++ software practices, solid algorithms, and parallelism, and resolve performance and correctness issues across chip and hardware subsystems.

What you'd actually do

  1. Drive architectural specifications to closure across multiple stakeholders.
  2. Develop written specifications for key component-level and system-level designs.
  3. Embody specifications in an executable model used by many customers across NVIDIA.
  4. Ensure high performance using good C++ software practices, solid algorithms and data structures, and parallelism.
  5. Resolve performance and correctness issues across chip and hardware subsystems by working across teams.

Skills

Required

  • BS or higher degree in a relevant field (CS, EE, Math) or equivalent experience
  • 5+ years of relevant experience
  • Expert programming and software skills in C++
  • Experience in RTL design and/or architecture
  • Ability to understand common chip design concepts
  • Automation-centered mindset

Nice to have

  • Experience with systems-level/architectural/chip modeling, profiling, and analysis
  • Strong combined RTL and C++ skillset

What the JD emphasized

  • LLM inference hardware
  • C++ hardware modeling experience

Other signals

  • LLM inference hardware
  • optimize AI inference speed and efficiency
  • executable model used by many customers