Robotics Engineer
ROBOSPEC PTE. LTD.
We are looking for a hands-on Robotics Engineerwith a passion for bringing AI to physical hardware. In this role, you will beat the core of our Embodied AI (Physical AI) pipeline, focusing ontransforming intelligent models into real-world robot actions.
Your primary responsibilitywill be programming, control, and real-world deployment on physicalmanipulator arms (such as Franka or UR). If you love working with realhardware, setting up cameras, and making mechanical arms move precisely andintelligently in the physical world, this is the role for you.
Key Responsibilities
● Real-Arm Deployment & Control:Design, implement, and maintain ROS 2-based control systems directly deployedon physical collaborative robots (cobots), specifically Franka, UR, orsimilar robotic arms.
● Embodied AI & Policy Integration:Develop core robot manipulation and behavior components using C++ and Python.Collaborate with the AI team to deploy VLA/VLM (Vision-Language-Action)models and imitation learning policies on real hardware.
● Full-Stack Robotics Integration:Integrate perception (e.g., cameras, sensors), manipulation primitives, andcontrol modules into a unified, reliable real-world system.
● Hardware Debugging & Optimization:Debug, profile, and optimize low-level robot control loop performance, addressreal-world latency, and solve sim-to-real gaps directly on the hardware.
● Simulation Support (Secondary): Utilizesimulation tools (e.g., Isaac Sim/Gazebo) as needed for basic policy trainingor initial behavioral testing before full-scale real-world deployment.
Required Skills & Qualifications
● Education: Bachelor’s degree orhigher in Robotics, Computer Science, Computer Engineering, MechanicalEngineering, or a related technical discipline. (Strong personalportfolio/projects will be highly valued).
● Hands-on Robot Experience: Strong, proventrack record of programming and controlling real robotic arms (Franka,Universal Robots, etc.).
● Robotics Frameworks: Deep experience withROS 2 and a solid foundation in robotics fundamentals (kinematics,dynamics, control paths, trajectories, and sensing).
● Programming: High proficiency in Pythonfor real-time or near-real-time applications.
Preferred (Bonus) Skills
● Robot Learning: Experience with imitationlearning (e.g., Diffusion Policy, ACT), reinforcement learning, or behaviorcloning applied to physical robots.
● Sim-to-Real Expert: Prior experiencenavigating the challenges of sim-to-real transfer, including domainrandomization or gap adaptation.
● End-to-End Delivery: Experiencedelivering end-to-end robotics applications (e.g., advanced picking, assembly,or machine apprenticeship tasks).