Field AI
Robotics Applied Scientist, Spatio-Temporal Scene Understanding & Memory
Irvine, CA · On-site · Full-time · Posted Oct 19, 2026
- Education
- PhD in robotics, computer science, electrical engineering, aerospace engineering, mechanical engineering, or a related field.
About the job
Develop persistent spatio-temporal mission memory and world models so robots can understand changing environments, reason and plan over past observations, and support long-term autonomy and physics-aware digital twins.
What you'll do
- Investigate representations for tracking what robots observe, where and when they observe it, and how environments change across missions and sites.
- Build world models for reasoning, planning, and recovering physical parameters through interaction.
- Combine learned and physically grounded models, using geometry and dynamics as ground truth.
What they ask for
- PhD in robotics, computer science, electrical engineering, aerospace engineering, mechanical engineering, or a related field, with relevant publications at top venues.
- Hands-on experience applying vision-language or foundation models to spatial reasoning, memory, or long-horizon embodied question answering.
- Depth in 3D representations and the estimation underlying them.
- Strong Python; ability to run methods on real robots with C++ and ROS 2.
Nice to have
- Change detection or long-term mapping results from real deployments; Real2Sim, system identification through interaction, or digital-twin work.
- Experience with behavior trees or language-model-driven task planning; industrial or field robotics exposure.
Tools
Python, C++, ROS 2
AI focus
Develop embodied AI methods for persistent scene memory, spatial reasoning, world modeling, and long-term robotic autonomy.
Model-written summary · the original posting is authoritative