Waymo is seeking to develop machine learning solutions to address open problems in autonomous driving, with the goal of safely operating Waymo vehicles in dozens of cities and under all driving conditions.
Requirements
Proficiency in writing and debugging Python/numpy-style code
Proficiency and in-depth knowledge of the inner workings of an ML framework (e.g. Pytorch, JAX, Tensorflow)
Proficiency in system performance: parallelism, buffering/prefetching and pipelining, and asynchronicity
Experience with large scale system reliability and failure-resilience/recovery
Expertise in ML accelerator programming, including but not limited to CUDA, Triton, or ML compilers
Knowledge of hardware micro-architectures and instruction sets
Proficiency in C++
Responsibilities
Enable the team to develop large scale end-to-end driving models with high hardware efficiency, reliable training and inference, and good developer experience
Participate in model design from a hardware-efficiency's perspective
Partner with other Waymo and Google teams to collaborate on infra work
Other
Bachelor degree in Computer Science, similar technical field of study, or equivalent practical experience
