NVIDIA is looking to develop its AI platform and specifically its performance optimizations in deep learning frameworks using JAX, aiming to deliver a polished, fast, modular, and coordinated platform for data, training, and analysis for a wide range of deep learning solutions.
Requirements
- C/C++ and Python programming
- Experience with machine learning frameworks and their internals (e.g. PyTorch, TensorFlow, scikit-learn, etc.)
- 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.
Responsibilities
- Play meaningful role in NVIDIA's effort in contributing to JAX.
- Design and implement JAX core components and drive peak performance on NVIDIA products.
- Work with AI applied researchers and leaders to build future-proof models
- Build tools that will increase the efficiency of teams developing AI-based systems.
- Work to bridge the gap between the latest in numerical computing, simulation and deep learning research and their applications in real world products.
Other
- 6+ years relevant experience
- Proven ability developing customer-facing solutions, balancing feature requests and bugs.
- Highly motivated with excellent verbal and written communication skills.
- Ability to work successfully with multi-functional teams, principles and architects. Coordinates effectively across organizational boundaries and geographies.
- BS in Computer Science or Computer Engineering or related field (or equivalent experience)
