Humanoid Robotic Engineer

Hyphen Connect Limited; Hong Kong; On-site; HKD 480,000 - 650,000; Full time

Job summary

Hyphen Connect Limited in Hong Kong is seeking a Humanoid Robotic Engineer specializing in bipedal locomotion and manipulation. This role involves designing sophisticated control algorithms, focusing on Reinforcement Learning to enhance robotic systems.

The engineer will develop advanced algorithms, train RL policies in simulations, and analyze real-world performance. A strong background in Model Predictive Control and experience with robotic simulation environments are essential.

Qualifications

• Strong background in Model Predictive Control (MPC) and Whole-Body Control.
• Extensive experience with robotic simulation environments.
• Familiarity with modern RL frameworks.

Responsibilities

• Develop and implement advanced control algorithms for bipedal locomotion and manipulation.
• Train Reinforcement Learning policies in simulation and deploy them on physical hardware.
• Conduct sim-to-real transfer and analyze gait/kinematics telemetry.

Skills

Model Predictive Control (MPC)

Whole-Body Control

Robotic simulation environments (e.g., NVIDIA Isaac Sim, MuJoCo)

Reinforcement Learning frameworks

Job description

We are searching for a Humanoid Robotic Engineer specialized in bipedal locomotion and manipulation. This role involves designing and implementing sophisticated control algorithms, with a focus on Reinforcement Learning, to enhance the performance of robotic systems both in simulation and real-world applications.

Responsibilities

• Develop and implement advanced control algorithms for bipedal locomotion and manipulation.
• Train Reinforcement Learning (RL) policies in simulation and deploy them on physical hardware.
• Conduct sim-to-real transfer and analyze gait/kinematics telemetry.

Required Skills

• Strong background in Model Predictive Control (MPC) and Whole-Body Control.
• Extensive experience with robotic simulation environments (e.g., NVIDIA Isaac Sim, MuJoCo).
• Familiarity with modern RL frameworks.