Senior System Simulation Architect

NVIDIA NVIDIA · Semiconductors · Santa Clara, CA +3

This role focuses on developing and integrating full-system functional models for CPU architectures, with a specific emphasis on enabling complex AI/DL and HPC workloads. The architect will work with simulation, emulation, and performance models to support the development of next-generation CPU products that are critical for NVIDIA's accelerated computing ecosystem.

What you'd actually do

  1. Develop full-system functional models capable of running complex multi-threaded heterogeneous (CPU/GPU) workloads – with special focus on the CPU subsystem.
  2. Integrate functional models from various frameworks with RTL simulators and emulators, hardware (HW-in-the-loop), and detailed performance models.
  3. Bring up system and application software in simulation and emulation – including firmware, Linux, drivers, benchmarks, and CPU/GPU workloads such as deep-learning (DL) and high-performance computing (HPC) workloads.
  4. Port/extend/develop system software (firmware, OS, and drivers) to meet workload simulation needs.
  5. Support CPU architects and performance engineers in their use of functional models, performance models, and emulation to drive next-generation CPU architectures.

Skills

Required

  • BS/MS in EE, CE, or CS or equivalent experience
  • 6 or more years of relevant experience
  • Excellent C/C++/Python programming skills
  • Experience in development of functional simulators and/or low-level software (OS, firmware, drivers); preferably both
  • Excellent debugging skills – of both system software/firmware and application software
  • Experience with the ARM ISA
  • Excellent communication and teamwork skills

Nice to have

  • Experience working with hardware emulators and/or FPGAs
  • Background in CPU workload analysis (SimPoint, etc.)
  • Experience with Linux kernel bringup and debug
  • Familiarity with CUDA
  • Experience with CPU/GPU application development and optimization in Pytorch, TensorFlow, and similar frameworks

What the JD emphasized

  • CPU architectures to fuel the explosive growth in artificial intelligence (AI) / deep learning (DL)
  • deep-learning (DL) and high-performance computing (HPC) workloads
  • CPU/GPU application development and optimization in Pytorch, TensorFlow, and similar frameworks