Robotics ML Engineer – Manipulation & Embodied AI

Keywords:Learning from Demonstration (LfD) / Imitation Learning / Reinforcement Learning / Diffusion Policy / VLA / Embodied AI / Robot Learning / Isaac Sim / Isaac Lab / Sim-to-Real / Manipulation / ROS2About the CompanyTrubot Inc. is a robotics company focused on intelligent inspection robots and automation solutions for data center operations.We build and deploy real robotic systems in industrial environments and are actively developing next-generation robotic manipulation capabilities for real-world automation tasks.Our long-term focus includes robotic manipulation, embodied AI, learning-based autonomy, and intelligent robotic systems operating in large-scale industrial environments.Role SummaryWe are seeking a Robotics ML Engineer to develop learning-based robotic manipulation systems for real-world deployment.This role focuses on robotic arms, robot learning, and embodied AI technologies that enable robots to acquire new skills from demonstration, simulation, and real-world interaction.The ideal candidate combines strong machine learning expertise with hands-on robotics experience and is excited about bringing advanced AI methods onto physical robotic platforms.ResponsibilitiesRobot Learning & ManipulationDesign, develop, and optimize learning-based robotic manipulation policiesDevelop systems for grasping, insertion, tool usage, and other real-world manipulation tasksIntegrate learned policies onto physical robot arms and mobile manipulation platformsCollaborate with robotics engineers to deploy manipulation systems in production environmentsLearning AlgorithmsDevelop and improve imitation learning, behavior cloning, reinforcement learning, and diffusion-based policiesExplore VLA (Vision-Language-Action), Embodied AI, and foundation-model-based approaches for robotic manipulationEvaluate and improve policy robustness, task success rates, and generalization performanceAnalyze failure modes and continuously improve system reliabilitySimulation & DataBuild and maintain simulation environments using Isaac Sim, Isaac Lab, MuJoCo, or similar platformsDevelop data collection pipelines through teleoperation and human demonstrationsManage training datasets and evaluation workflowsSupport sim-to-real transfer and deployment onto physical robotic systemsReal-World DeploymentDeploy learned policies onto real robotic armsDebug and resolve sim-to-real performance gapsWork closely with hardware, sensing, and software teamsParticipate in system integration and field validationRequired QualificationsMS or PhD in Robotics, Computer Science, Artificial Intelligence, Machine Learning, Electrical Engineering, or a related fieldStrong programming skills in Python and/or C++Experience with ROS2 and robotic system integrationExperience with robotic manipulation systemsExperience with machine learning for robotics applicationsFamiliarity with Linux development environmentsPreferred QualificationsExperience in one or more of the following areas:Learning from Demonstration (LfD)Imitation Learning (IL)Behavior CloningReinforcement Learning (RL)Diffusion PolicyACT (Action Chunking Transformer)VLA (Vision-Language-Action)OpenVLARobot Foundation ModelsIsaac SimIsaac LabMuJoCoSim-to-Real TransferTeleoperation SystemsReal-World Robot DeploymentMulti-Modal Robot LearningVision-Language Models for RoboticsEmbodied AIWhat We ValueWe are not looking for candidates who only publish papers or only work in simulation.We highly value engineers who can bridge research and deployment, turning advanced robot learning algorithms into reliable real-world robotic systems.Experience deploying robotic manipulation policies on physical robots is strongly preferred.LocationFremont, California (Onsite)ApplicationWe welcome applications from qualified candidates.Please include "Robotics ML Engineer" in the email subject line.Please submit your resume to:info@trubot.ai