The Active Prediction Integration team is responsible for the development of a unified prediction and planning architecture, by integrating ML-based and first-principles-based models. It is a highly cross-functional team that collaborates with planning, prediction and behavior ML, and perception engineers to critically influence robot behaviors. You'll be part of a passionate team dedicated to enabling Zoox's rider-first experience by making the autonomous robot safer, smarter and more comfortable.
Together with the team, you will advance the state of the art in prediction and decision-making. We are looking for engineers with experience in developing and integrating advanced ML models in safety-critical production environments. If you enjoy learning new AI technologies and embrace the challenges of evolving prediction and planner architectures, we would like to hear from you.
Responsibilities
You will work closely with Prediction and Planner to integrate advanced ML models and develop first-principles based safeguards, to ensure trustworthy prediction and decision making.
You will inform and guide the development of new ML models and should be willing to dive deep and address issues in the model and data pipelines.
You will design interfaces between prediction and planning, creating abstraction layers that facilitate modular integration of various machine learned components within the planning architecture.
You will create principled algorithmic improvements to specific areas of driving behavior and implement those changes in our code base.
You will facilitate the collaborations between the prediction data/modeling teams and the planning and control teams in order to deliver architectural changes to the system.
Qualifications
Master's or PhD degree in computer science or related field and 5+ years of relevant experience
Fluency in modern C++ standards, programming and software design principles
Strong knowledge of algorithmic design, implementation and optimization
Prior experience or expertise in machine learning, either in model architecture or production data pipelines
Demonstrated ability to communicate and collaborate cross-functionally
Bonus Qualifications
Demonstrated ability to create real-time motion planning algorithms
Experience with sequential decision making, including, but not limited to, imitation learning, reinforcement learning, joint trajectory prediction and planning, etc.
Experience with machine learning and deep learning frameworks (e.g., TensorFlow, PyTorch, TensorRT) and basic knowledge of CUDA programming or willingness to learn
Compensation
There are three major components to compensation for this position: salary, Amazon Restricted Stock Units (RSUs), and Zoox Stock Appreciation Rights. The salary will range from $221,000 to $319,000. A sign-on bonus may be part of a compensation package. Compensation will vary based on geographic location, job-related knowledge, skills, and experience.
Zoox also offers a comprehensive package of benefits including paid time off (e.g. sick leave, vacation, bereavement), unpaid time off, Zoox Stock Appreciation Rights, Amazon RSUs, health insurance, long-term care insurance, long-term and short-term disability insurance, and life insurance.