Zoox is developing the first ground-up, fully autonomous vehicle fleet and the supporting ecosystem required to bring this technology to market. The Perception Object Detection and Tracking team at Zoox deals with perception of all people and objects that have a capability to move. Your role is to work with the ML model teams to bring cutting-edge models into the vehicle stack.
Requirements
- Fluency in C++ and Python
- Experience delivering ML model integration in latency-sensitive systems
- Have experience implementing tracking systems
- Strong mathematical skills and understanding of probabilistic techniques
- Familiarity with modern, Sparse-BEV Joint Detection and Tracking models
- Experience with CUDA code
Responsibilities
- Define on-vehicle architecture for producing the core tracking results from the Perception stack.
- Work with both the model teams and optimization teams to develop a highly performant and efficient system that can run on vehicle.
- Work with Perception data, both on the input and output of machine learned models
- Take tracking output and integrate this into the larger behavioral system in the Autonomy stack
- creating new algorithms for segmentation, tracking, classification, and high-level scene understanding
- Investigate, prototype and train / evaluate networks for solving causal / recursive multi-target multi-modal tracking problem
Other
- BS or higher in Computer Science or a related degree
- Contribute to the ongoing development of new architecture based on new, state of the art ML research
- You do not need to match every listed expectation to apply for this position.
- We are looking for top talent that shares our passion and wants to be part of a fast-moving and highly execution-oriented team.
