Robotics Motion Control Algorithm Intern

About the role

Join BotsDeploy’s research team to advance robotics and AI platforms. As a Robotics Motion Control Algorithm Intern you will design and test intelligent control policies that enable bimanual manipulation and whole‑body locomotion on humanoid robots.

Key responsibilities

• Develop control policies for bimanual manipulation and whole‑body coordination using Reinforcement Learning, Imitation Learning and Diffusion Policy methods.
• Enable complex behaviours such as object carrying, tool use, walking, stair climbing and obstacle traversal.
• Run large‑scale training and evaluation on high‑performance simulators (e.g., Isaac Lab, MuJoCo) and investigate sim‑to‑real transfer techniques.
• Deploy trained policies onto real humanoid robots and assess performance.
• Research multi‑task, multi‑modal and multi‑robot shared policy training to improve generalisation and stability.
• Integrate vision‑language‑action models for task understanding, decision‑making and closed‑loop control.

Required profile

• Major in Computer Science, Artificial Intelligence, Mathematics, Robotics or a related field.
• Strong coding abilities and solid algorithmic foundations.
• Good understanding of robot kinematics and dynamics.
• Familiarity with robot reinforcement‑learning techniques and hands‑on experience with real‑world implementation.

Required skills

• Python
• C++
• Linux operating systems
• Reinforcement‑learning algorithms (PPO, SAC, TD3, Dreamer, GAIL)
• Imitation‑learning methods
• Diffusion‑based policies (Diffusion Policy, Guided Diffusion, Flow Matching)
• Trajectory optimisation
• Robot kinematics and dynamics
• Vision‑Language‑Action model integration