Senior Manager, Deep Learning Performance Architecture

NVIDIA NVIDIA · Semiconductors · Shanghai, China +1

NVIDIA is seeking an Engineering Manager to lead a Deep Learning Performance Architect Team. This role involves managing a team focused on analyzing deep learning networks and advancing deep learning computing systems through hardware/software co-design. Responsibilities include establishing team objectives, collaborating with software framework and hardware architecture teams, characterizing deep learning workloads, performance tuning, optimizing software stacks, and driving the evolution of next-generation hardware and software architectures.

What you'd actually do

  1. managing a team of experienced deep learning performance architects to analyze deep learning networks and push the evolution of our deep learning computing system in hardware/software co-design approach
  2. establish team objectives to meet schedules and goals
  3. establish and evolve policies and procedures that affect the immediate organization
  4. communicate with senior management for team vision and development
  5. collaborate with members of the deep learning software framework teams and the hardware architecture teams to accelerate the next generation of deep learning computing system

Skills

Required

  • Bachelors, Masters or Ph.D. or equivalent in Computer Science, Computer Engineering, related field (or equivalent experience)
  • 10+ years of overall experience with at least 6+ of those years with hands on management
  • Strong software design fundamentals
  • deep understanding of deep learning optimization
  • Knowledge of software engineering principles

Nice to have

  • Knowledge of CPU and/or GPU architecture
  • CUDA or OpenCL programming experience
  • Experience with XLA, TVM, MLIR, LLVM, deep learning models and algorithms, and deep learning framework design
  • Previous work on large complex codebases with many other developers, especially libraries, compilers, or system software
  • Track record of identifying new technologies and incorporating them into software development flows
  • Excellent understanding of linear algebra and calculus

What the JD emphasized

  • deep learning optimization
  • deep learning models and algorithms

Other signals

  • deep learning computing system
  • hardware/software co-design
  • deep learning workloads characterization
  • performance tuning and analysis
  • optimizing the present generation of our software tech stack
  • drive the evolution of the next generation of deep learning hardware and software architecture