About the job
Join our team in developing sophisticated performance optimizations for deep learning frameworks using JAX, a powerful tool that combines automatic differentiation with XLA for high-performance machine learning research. We are committed to creating a fast, modular, and coordinated platform for a wide range of deep learning solutions. Together, we will push the boundaries of AI and scientific computing to new frontiers! If you're a skilled programmer with a knack for elegant design and a passion for developing customer-facing solutions, this is your opportunity to make a lasting impact on our groundbreaking work. We're looking for highly motivated individuals with excellent verbal and written communication skills who are ready to tackle complex challenges in AI and scientific computing.
Responsibilities
Make meaningful contributions to JAX by designing and implementing core components that drive phenomenal performance on the NVIDIA AI platform.
Increase the efficiency of teams developing JAX-based systems by develop tools that streamline the application of deep learning research and simulations to real-world products.
Qualifications
Minimum
Pursuing a MS or PhD in Computer Science, Computer Engineering or equivalent program area.
Understanding of JAX, Autograd, tracing, code generation and DSL compilers
Working proficiency in Python, familiarity with C++,
Knowledge of machine learning frameworks (e.g., PyTorch, TensorFlow, scikit-learn)
Preferred
Familiar with distributed systems, services, or Deep Learning at large scale
Ability to work optimally with multi-functional teams
Proven technical foundation in CPU and GPU architectures, numeric libraries, modular software design