Software Engineer, ML Platform (ML Training)

Zoox•Foster City, CA

2d•$182,000 - $257,000

About The Position

The ML Platform team at Zoox plays a crucial role in enabling innovations in ML and CV to make autonomous driving as seamless as possible. The ML Training platform team builds and operates the core part of the ML platform that powers model training at scale. We are responsible for developing and operating ML tools, deep learning frameworks, and distributed model training infrastructure to support foundational models and reinforcement learning. This team also owns the model repository and model lifecycle management tools used by our applied research teams for in- and off-vehicle ML use cases. You will play a crucial role in reducing the time it takes from ideation to productionization of cutting-edge AI innovation. This team has a lot of growth opportunities as we expand our robotaxi deployments and venture into new ML domains.

Requirements

2+ years of ML infrastructure experience.
Experience with training frameworks like PyTorch, DeepSpeed, JAX, Ray, etc.
Experience working with cloud providers like AWS.

Nice To Haves

Experience with building large-scale, cost-efficient distributed model training and ML compute infrastructure.
Experience with building model lifecycle management tools and experimentation.

Responsibilities

Build the Zoox Training framework leveraged by all ML teams within Zoox. This framework needs to be highly scalable, reliable, and efficient.
Design, implement, and operate a robust and efficient ML platform to enable the training, validation, serving, and monitoring of ML models.
Collaborate closely with cross-functional teams, including ML researchers, software engineers, and data engineers, to define requirements and align on architectural decisions.