Description

You will design and implement advanced robot learning architectures to support dexterous manipulation, path planning, and autonomous task sequencing. You will develop end-to-end policy training pipelines that integrate multi-modal sensory data—including RGB, depth, proprioception, force/torque, LiDAR, and tactile inputs—with control outputs. You will build policy inference and closed-loop control systems connecting perception, planning, and execution on physical robotic platforms. You will apply and extend large-scale architectures such as LLMs, VLMs, and VLAs to embodied tasks, grounding, and sim-to-real adaptation. You will lead data strategy for demonstrations, teleoperation, simulation pipelines, and evaluation frameworks. You will collaborate with cross-functional teams to deploy robot policies on hardware, ensuring robustness, repeatability, and safety. You will stay current with embodied AI research and share insights through mentorship and technical presentations.

Responsibilities

• Design advanced robot learning architectures including diffusion policies, ACT, and imitation learning.
• Develop end-to-end policy training pipelines integrating multi-modal sensory data with control outputs.
• Build policy inference and closed-loop control systems for physical robotic platforms.
• Apply large-scale architectures (LLMs, VLMs, VLAs) to embodied tasks and sim-to-real adaptation.
• Lead data strategy for demonstrations, teleoperation, and evaluation frameworks.

Required Skills

• 6+ years of experience in applied science or machine learning.
• Master's degree in a relevant technical field.
• Expertise in Python and AI/ML frameworks.
• Deep knowledge of LLMs and VLMs/VLAs.
• Experience with robot learning, imitation learning, and diffusion policies.
• Strong background in multi-modal sensor data integration.
• Experience with sim-to-real adaptation and embodied AI research.

Key Skills

Education

Master's degree