The Perception HAL team makes civilized Perception possible! We provide the stable sensing stack on which the Perception models run, which means providing and rendering sensor data (including fusion), calibrating and validating sensors, estimating weather and visibility, and making the whole stack deal gracefully with degraded sensing or failures. We interface with the Waymo HW teams and have huge influence on the HW design of new sensors. We're looking for a lead to work on the architecture of Perception's alternate pipelines that are used when the main Perception stack can no longer be run.

You will design the architecture and design of this stack, guide the work of engineers across the entire Perception team to implement it, and collaborate inside a larger, Waymo-wide working group to deliver an end-to-end solution.

In this hybrid role you will report to an Engineering Director

You will:

• Be involved with the state of the art in ML model development in Perception (these projects are often where we can try our new tech)
• Develop technical leadership across the whole Perception stack, and also well-positioned to collaborate beyond the team with hardware and behavior (planner) teams and get great full-stack experience

You have:

• M.S. in Computer Science (or related fields) with 10 years work experience
• 5+ years experience in developing, optimizing, evaluating, and deploying ML algorithms into large-scale software systems.
• Academic background and industry experience in one or more related fields: Machine Learning, Computer Vision, Computer Graphics & Rendering;

We Prefer:

• Ph.D. in Computer Science (or related fields) and 7 years work experience;
• 3+ years' experience in simulation, perception, autonomous driving, and robotics
• Direct experience using or developing multi-modal foundation models
• Background in perception related domains, such as transformer modeling, model scaling, self-supervised learning
• Publications at top-tier conferences like CVPR, ICCV, ECCV, ICLR, ICML, ICRA, RSS, NeurIPS, AAAI, IJCV, PAMI

#LI-Hybrid