Distinguished Engineer, Jax

NVIDIA · Semiconductors · Santa Clara, CA +1 · Remote

Distinguished Engineer to develop NVIDIA's AI platform, focusing on performance optimizations in deep learning frameworks using JAX. The role involves designing and implementing core JAX components, driving peak performance on NVIDIA products, and building tools to increase the efficiency of AI-based system development teams. It bridges numerical computing, simulation, and deep learning research with real-world applications.

What you'd actually do

  1. Play significant role in NVIDIA's effort in contributing to JAX.
  2. Design and implement JAX core components and drive peak performance on NVIDIA products.
  3. Work with AI applied researchers and leaders to build future-proof models
  4. Build tools that will increase the efficiency of teams developing AI-based systems.
  5. Work to bridge the gap between the latest in numerical computing, simulation and deep learning research and their applications in real world products.

Skills

Required

  • C/C++ and Python programming
  • Experience with machine learning frameworks and their internals (e.g. PyTorch, TensorFlow, scikit-learn, etc.)
  • Proven ability developing customer-facing solutions, balancing feature requests and bugs.
  • Proven technical foundation in CPU and GPU architectures, numeric libraries, modular software design.
  • Excellent communication and planning skills
  • Ability to work successfully with multi-functional teams, principles and architects. Coordinates effectively across organizational boundaries and geographies.

Nice to have

  • Understanding of JAX, Autograd, tracing, code generation and DSL compilers and their design.
  • Understanding of deep learning training in distributed contexts: multi-GPU, multi-node, synchronous vs asynchronous.
  • Background with software shipping cycles (dev, deploy, release, CI).
  • Experience building distributed systems and services at large scale.

What the JD emphasized

  • 18+ years relevant experience
  • Proven ability developing customer-facing solutions, balancing feature requests and bugs.
  • Proven technical foundation in CPU and GPU architectures, numeric libraries, modular software design.
  • Understanding of JAX, Autograd, tracing, code generation and DSL compilers and their design.
  • Understanding of deep learning training in distributed contexts: multi-GPU, multi-node, synchronous vs asynchronous.
  • Background with software shipping cycles (dev, deploy, release, CI).
  • Experience building distributed systems and services at large scale.

Other signals

  • performance optimizations
  • deep learning frameworks
  • JAX
  • numerical computing
  • machine learning research