Engineer, Embodied AI

About the Role

We seek an innovative Embodied AI and Robotics Intelligence Research & Development Engineer to create advanced robot intelligence systems. This role emphasizes integrating large AI models with real-world robotics for perception, planning, decision-making, and control. The engineer will design modular robot brain architectures linking LLM, VLM, VLA, spatial intelligence, world models, and task execution based on ROS 2.

Key Responsibilities

• Design and implement embodied AI algorithms to enable robotic perception, reasoning, task planning, and autonomous operations.
• Investigate and integrate large language, vision, and audio models into robotic platforms operating in real environments.
• Develop multi-layered robot planning frameworks covering LLM-driven planning, task and skill planning, and execution.
• Create and maintain robot skill and capability representations, task graphs, and libraries of reusable skills.
• Conduct research on world modeling, spatial cognition, and persistent memory for enabling long-duration robotic activities.
• Generate semantic-geometric representations that combine visual semantics, 3D spatial data, and robot/environment states.
• Perform simulations and physical experiments on various robotic platforms including mobile bases, quadrupeds, and manipulators.
• Stay up to date with advancements in embodied AI, robot foundation models, VLA, world models, and autonomous robotics.

Required Qualifications

• Master's degree or higher in Robotics, AI, Computer Science, Automation, Electrical or Mechanical Engineering, or closely related fields.
• Proficient programming skills in Python and/or C++.
• Strong foundational knowledge in robotics topics such as robot perception, task planning, simultaneous localization and mapping (SLAM)/navigation, manipulation, and reinforcement learning.
• Experience working with ROS and ROS 2 frameworks.
• Familiarity with contemporary deep learning architectures including Transformers, Vision-Language Models (VLM), Large Language Models (LLM), or generative models.
• Capability to autonomously implement and test research concepts on physical robot platforms.

Additional Advantages

Background in actual robot deployments, published research, patents, or contributions to open-source robotics projects will be considered a significant asset.