Senior Deep Learning Performance Architect

NVIDIA · Semiconductors · Santa Clara, CA +1

Senior Deep Learning Performance Architect role at NVIDIA focused on developing and analyzing next-generation architectures for AI and HPC applications. This involves performance modeling, simulation, and understanding the interplay of hardware and software for deep learning training and inference.

What you'd actually do

  1. Develop innovative architectures to extend the state of the art in deep learning performance and efficiency
  2. Analyze performance, cost and power trade-offs by developing analytical models, simulators and test suites
  3. Understand and analyze the interplay of hardware and software architectures on future algorithms, programming models and applications
  4. Develop, analyze, and harness groundbreaking Deep Learning frameworks, libraries, and compilers
  5. Actively collaborate with software, product and research teams to guide the direction of deep learning HW and SW

Skills

Required

  • GPU or Deep Learning ASIC architecture for training and/or inference
  • performance modeling
  • architecture simulation
  • profiling
  • analysis
  • machine learning and deep learning
  • Python
  • C
  • C++

Nice to have

  • deep neural network training, inference and optimization in leading frameworks (e.g. Pytorch, JAX, TensorRT)
  • relevant libraries, compilers, and languages - CUDNN, CUBLAS, CUTLASS, MLIR, Triton, CUDA, OpenCL
  • architecture of or workload analysis on other DL accelerators
  • self-motivation
  • critical thinking
  • thinking outside the box

What the JD emphasized

  • performance analysis
  • performance modeling
  • AI/deep learning
  • deep learning performance and efficiency
  • analytical models
  • simulators
  • test suites
  • hardware and software architectures
  • Deep Learning frameworks, libraries, and compilers
  • deep learning HW and SW
  • GPU or Deep Learning ASIC architecture
  • training and/or inference
  • performance modeling
  • architecture simulation
  • profiling
  • analysis
  • machine learning and deep learning
  • deep neural network training, inference and optimization
  • leading frameworks
  • relevant libraries, compilers, and languages

Other signals

  • performance analysis
  • architecture development
  • deep learning efficiency
  • hardware/software co-design