Within the Motion Planning organization at Zoox, you’ll be dedicated to improving the driving behaviors of the robot on public roads. The Planner's job is to figure out where the robot is going and how to get there safely in situations as complex as those found in Downtown San Francisco or on the Las Vegas Strip.
The Trajectory Generation team is responsible for developing robust and efficient algorithms and software that produces safe, feasible, and optimal trajectories. This module is a key component of the Planner for our bidirectional autonomous vehicle, and it is critical to achieving a safe and comfortable rider experience in various driving environments. We are seeking an engineer to develop and enhance our trajectory generation algorithms and software to continuously improve our autonomous driving system.
In this role, you will:
Design trajectory generation algorithms and implement them in high-quality software to continuously improve the capability of our autonomous driving system and enhance the safety and comfort of our driving behavior on public roads.
Use our extensive backend tooling to test and validate your algorithms and software across many different driving situations, and develop metrics and pipelines to further our ability to test/measure the algorithms’ performance.
Participate in or lead cross-functional projects to deliver mission-critical features to expand our autonomous driving capabilities.
Participate in on-vehicle drive reviews to gather actionable feedback on driving performance, and use this knowledge to help the team learn and make continuous improvements.
Conduct literature review and participate in study groups to stay current with the advancements in the fields of Motion Planning, Autonomous Driving, and Robotics in general.
Qualifications
Master's degree in Computer Science, Robotics, or related fields
Fluency in C++ with experience shipping productionized software
Proven experience in developing trajectory optimization algorithms and real-time motion planning software
Deep knowledge about numerical constrained optimization methods
Bonus Qualifications
Experience with geometric, sampling, and search-based planning algorithms
Experience with machine learning-based motion planning
Experience with behavior planning, discrete decision making, tree search, and POMDP
