Dexterous Robotics Research Intern

Meta; Redmond (WA); On-site; USD 55,000 - 83,000; Full time

Job summary

Agios in Reality Labs Research is seeking a Research Scientist Intern to explore robot learning, dexterous control, and human-to-robot transfer. You will develop learned models from human demonstrations and contribute to teleoperation and dexterous manipulation studies.

You will work with mentors spanning hardware, simulation, controls, and policy learning, shaping data-driven methods for advanced robotic Hands and real-world experiments.

Qualifications

• Currently pursuing or holding a Ph.D. in CS, AI, Robotics, or a related field.
• Strong understanding of dexterous manipulation and control.
• Experience with Python and PyTorch or JAX; familiarity with ROS or comparable robotics middleware.
• Proven record of significant results via grants, fellowships, patents, or leading publications at major venues.
• Experience training modern control policies including flow matching, diffusion, imitation learning, and RL.
• Experience with robotics modeling including kinematics, dynamics, trajectory optimization, planning, perception, and control theory.
• Hands-on experience with multi-fingered robotic hands and grippers on a robot arm.
• Experience with sim-to-real and real-to-sim workflows; domain randomization, system identification.
• Ability to communicate across functions and return to the degree program after internship.
• Availability for minimum 16 consecutive weeks.

Responsibilities

• Drive an independent research thread on learned control and motion transfer for dexterous hands, from problem formulation through evaluation.
• Build training pipelines for modern policy classes, including flow matching, diffusion policies, imitation learning, RL, large behavior models, and VLAs.
• Develop representations that fuse multimodal human and robot data into a shared action space.
• Construct data collection pipelines and benchmarks for dexterous manipulation across objects and embodiments.
• Collaborate across hardware, simulation, controls, and policy researchers, and communicate plans, progress, and results.

Skills

Python

PyTorch

JAX

ROS

Dexterous manipulation knowledge

Reinforcement learning

Multimodal representation learning

Cross-embodiment transfer

Trajectory optimization

Data collection for robot demos

Sim-to-real workflows

Computer vision

Time-series sensor data

Hands-on robotics

Education

Ph.D. in Computer Science / AI / Robotics or related field

Tools

Python

PyTorch

JAX

ROS

Job description

Agios in Reality Labs Research is seeking a Research Scientist Intern to explore robot learning, dexterous control, and human-to-robot transfer. You will develop learned models from human demonstrations and contribute to teleoperation and dexterous manipulation studies.

You will work with mentors spanning hardware, simulation, controls, and policy learning, shaping data-driven methods for advanced robotic Hands and real-world experiments.