Robot Motion Planning & Prediction - ML Engineer

We are seeking an ML Engineer to join our robot motion planning team. In this role, you will build and deploy an ML-based motion planning stack that generates socially-aware, real-time trajectories for outdoor delivery robots operating in complex urban environments.

This work includes developing closed-loop behavioral models of surrounding agents and training and evaluating a safe ML-based planner on top of this simulation framework to ensure robust and predictable behavior in both simulation and real-world tests.

• Develop closed-loop behavioral models of agents using behavioral cloning and related learning techniques. Build a simulation framework that uses these models to generate realistic multi-agent interactions.
• Design, train, and evaluate ML-based motion planner that operate safely and efficiently in real-time environments. Define evaluation metrics and run large-scale experiments in both simulation and live-ride testing.
• Collaborate with planning, simulation, and perception teams to integrate your models into the full autonomy stack. 

• 3+ years of ML engineering experience or a PhD in a related field.
• Strong Python skills and experience with PyTorch.
• Knowledge of modern C++ and a solid understanding of high-performance code design.
• Solid understanding of machine learning fundamentals and ability to design, train, and evaluate ML models end-to-end — including data preparation, training pipelines, and validation. 

• PhD in Computer Science, Machine Learning, Robotics, or a related field.
• Experience in ML-based motion planning or related robotics problems.
• Experience with robotics simulation tools or custom data-driven simulators.

Candidates are required to be authorized to work in the U.S. The employer is not offering relocation sponsorship, and remote work options are not available.