Research Engineer (SRA/ARM), SIMTech

A*STAR - Agency for Science, Technology and Research; Singapore; On-site; SGD 60,000 - 90,000; Full time

Job summary

A*STAR - Agency for Science, Technology and Research in Singapore is hiring a Research Engineer/Senior Research Engineer to develop and implement humanoid and multi-arm robotic manipulation systems for industrial tasks. The role requires a Bachelor's or Master's degree in Robotics or a related field, and strong programming skills in Python and/or C++. You'll engage in designing multimodal perception systems, implementing learning-based manipulation policies, and testing robotic prototypes to enhance task success in manufacturing environments.

Qualifications

• Bachelor’s or Master’s degree in Robotics, Computer Engineering, Mechanical Engineering, Computer Science, Automation, or related fields.
• Strong programming skills in Python and/or C++, with experience building robotics software systems.
• Experience with robotics frameworks such as ROS and robot middleware for sensor and controller integration.

Responsibilities

• Design and develop multimodal perception systems for humanoid and multi-arm robots.
• Implement and deploy learning-based manipulation policies for coordinated tasks.
• Build, test, and refine humanoid and multi-arm robotic prototypes for industrial applications.

Skills

Programming skills in Python

Programming skills in C++

Robotics software system development

Multimodal perception systems

ROS frameworks

Education

Bachelor's or Master's degree in Robotics or related fields

Job description

Job Description

The mission of the Adaptive Robotics and Mechatronics (ARM) group in the Singapore Institute of Manufacturing Technology (SIMTech), Agency for Science, Technology and Research (A*STAR) is to advance the state of intelligent, adaptive, and embodied robotic and mechatronic systems for next-generation manufacturing.

Role Summary

We are seeking a Research Engineer/Senior Research Engineer to develop, implement, and deploy humanoid and multi-arm robotic manipulation systems for real-world industrial tasks. This role focuses on developing and transforming vision-based and learning-based robot manipulation into reliable, real-time humanoid and dual/multi-arm robotic systems capable of operating in complex, unstructured, and contact-rich environments. You will work at the intersection of perception, motion planning, and full-body control and manipulation, building integrated pipelines that allow humanoid/multi-arm robot to complete coordinated manipulation for tasks.

Responsibilities

• Design and develop multimodal perception systems for humanoid and multi-arm robots, integrating vision, force/tactile sensing to perceive environment and interaction for manipulation.
• Implement and deploy learning-based manipulation policies, including vision-language-action models, multimodal skill networks, and imitation-learned behaviours, enabling humanoid and dual-arm robots to perform coordinated bimanual and whole-body manipulation.
• Integrate perception, action models, and planning into real-time execution systems, ensuring stable, safe, and precise contact-rich manipulation under uncertainty and environmental variability.
• Build, test, and refine humanoid and multi-arm robotic prototypes, validating embodied AI pipelines on physical robots and improving robustness, repeatability, and task success in industrial and manufacturing environments.

Job Requirements

• Bachelor’s or Master’s degree in Robotics, Computer Engineering, Mechanical Engineering, Computer Science, Automation, or related fields.
• Strong programming skills in Python and/or C++, with experience building robotics software systems.
• Experience with robotics frameworks such as ROS and robot middleware for sensor and controller integration.
• Familiarity with robot kinematics, motion planning, and sensor calibration, especially for manipulation systems.
• Exposure to learning-based perception and robotic control, such as pose estimation, imitation learning, reinforcement learning, or multimodal policy models.