Senior Research Engineer – Visuomotor Robotics for Humanoids

ntu; Singapore; On-site; SGD 120,000 - 180,000; Full time

Job summary

NTU invites applications for a Senior Research Engineer with a strong robotics learning background to lead research on visuomotor manipulation policies for humanoid robots, advancing the state of the art and enabling real-world deployment.

The role involves defining research directions, developing learning-based manipulation techniques, and collaborating with perception, controls, systems, and hardware teams to integrate learned policies into the autonomous stack.

Qualifications

• PhD in Robotics, CS, EE, or related field.
• Strong publication record in robot learning and manipulation.
• Experience deploying robot learning on physical platforms.
• Expertise in visuomotor control and manipulation.
• Proficiency in Python and PyTorch.
• Ability to lead and mentor researchers.

Responsibilities

• Lead development of learning-based visuomotor policies for humanoid robot manipulation from concept to real-world validation.
• Define research directions and roadmaps for grasping, reorientation, bimanual manipulation, and assembly.
• Advance techniques such as behavior cloning, RL, and planning with VLA-based reasoning.
• Develop robust methods against sensor noise, partial observability, and dynamic contact.
• Oversee data collection, training, evaluation, and deployment pipelines.
• Coordinate with collaborators; report progress to the PI.
• Supervise research associates, engineers, and PhD students.
• Collaborate with perception, controls, systems, and hardware teams to integrate learned policies.
• Evaluate tradeoffs between learning-based and classical approaches.
• Lead preparation of publications, reports, and grant documentation.

Skills

Robot learning

Visuomotor control

Python

PyTorch

Publications

Leadership

Communication

Education

PhD in Robotics/CS/EE

Tools

PyTorch

Job description

NTU invites applications for a Senior Research Engineer with a strong robotics learning background to lead research on visuomotor manipulation policies for humanoid robots, advancing the state of the art and enabling real-world deployment.

The role involves defining research directions, developing learning-based manipulation techniques, and collaborating with perception, controls, systems, and hardware teams to integrate learned policies into the autonomous stack.