Robotics Applied Scientist, Active Perception

Description

The Irvine team works on embodied AI with real robots, real sensors and field deployments, based in Southern California's robotics hub.

The company builds AI systems that handle risk, stay reliable and are ready for field use, tackling some of robotics' toughest problems.

The team goes past typical data-driven methods and transformer-only designs, combining careful engineering with learning methods already running in products around the world.

Robots get better each time they run in the field, and the Irvine group tests its work on real hardware before systems ship to operate in unfamiliar environments worldwide.

The team includes AI, software, robotics engineering, product, field deployment and technical communication staff, and colleagues across the US and abroad.

Team members have come from DeepMind, NASA JPL, Boston Dynamics, NVIDIA, Amazon, Tesla Autopilot, Cruise, Zoox, Toyota Research Institute and SpaceX, and the company has field deployment history and DARPA challenge results.

Responsibilities

• Frame the active perception problem for inspection and a wide range of robot behaviors.
• Build viewpoint planning policies that gather information actively.
• Study calibrated confidence and abstention for vision-language reasoning on a robot.
• Take research and development through real-world experiments that show whether an approach is feasible and deployable.
• Help define the technical roadmap by matching real-world needs with development work.
• Mentor junior team members and collaborators.

Requirements

• A PhD in robotics, computer science, electrical engineering, aerospace engineering, mechanical engineering or a related field, with publications at venues such as RSS, CoRL, ICRA, IROS, T-FR, T-RO, CVPR, NeurIPS or similar.
• Research experience in active perception, informative path planning, vision-language models or semantic mapping for exploration.
• Hands-on work with vision-language or foundation models for robot perception, along with the physical principles behind them.
• Fluent Python, plus the ability to run your methods on real robots using C++ and ROS 2.
• Solid working knowledge of uncertainty, including Bayesian or information-theoretic planning, calibration and conformal methods.

Conditions

Based in Irvine, California.

The salary range is described as generous, and final pay depends on role scope, job knowledge, skills, experience and the Irvine market.

The company values diverse perspectives and judges candidates on merit, qualifications and performance, without regard to protected status.