Robotics Applied Scientist: Spatio-Temporal World Models

FieldAI; Irvine (CA); Vor Ort; USD 150.000 - 210.000; Vollzeit

Zusammenfassung

FieldAI’s Irvine team develops embodied AI systems for real robots using real hardware and deployments. You will advance memory representations, world models, and physics-aware digital twins to enable robust, long-term autonomy in challenging environments.

The role requires a PhD in robotics or related field, hands-on experience with vision-language models, and strong Python/C++ skills to run methods on real robots with ROS 2.

Qualifikationen

• PhD in robotics, computer science, electrical engineering, aerospace or related field, with publications in top venues.
• Hands-on experience with vision-language or foundation models applied to spatial reasoning, memory, or long-horizon embodied QA.
• Depth in 3D representations: scene graphs, neural/ Gaussian-splat maps, or object-centric maps.

Aufgaben

• Investigate representation for robot's mission memory (persistent spatio-temporal).
• Build world models enabling reasoning and planning over memory and physics-aware twins.
• Combine learned and physically grounded models with geometry and dynamics as ground truth.
• Bridge perception-reasoning gap for long-term autonomy.

Kenntnisse

Python

ROS 2

C++

Vision-language

Spatial reasoning

Ausbildung

PhD in robotics or related field

Tools

ROS 2

C++

Python

Jobbeschreibung

FieldAI’s Irvine team develops embodied AI systems for real robots using real hardware and deployments. You will advance memory representations, world models, and physics-aware digital twins to enable robust, long-term autonomy in challenging environments.

The role requires a PhD in robotics or related field, hands-on experience with vision-language models, and strong Python/C++ skills to run methods on real robots with ROS 2.