Principal Manipulation Engineer

Maxwell Bond; Reno (NV); On-site; USD 120,000 - 180,000; Full time

Job summary

Maxwell Bond is seeking a Robotics Manipulation Engineer to develop, implement, and optimize AI-driven manipulation for humanoid robots using reinforcement learning, diffusion and flow models, and vision-language-action systems.

You will bridge AI research with real-world deployment, building robust, scalable manipulation capabilities and coordinating with cross-disciplinary teams to transition from simulation to hardware.

Qualifications

• 2+ years of robotics or AI/manipulation experience.
• Strong programming in Python and/or C++.
• Experience with ML frameworks (PyTorch or TensorFlow).
• Experience training RL policies and reward design.
• Knowledge of ROS/ROS2 and real robot deployment.
• Solid grasp of robot kinematics and motion planning.

Responsibilities

• Develop manipulation algorithms for high-DOF humanoid hands.
• Train RL policies and generative models for manipulation.
• Integrate vision-language-action models into perception-to-action pipelines.
• Develop sim-to-real pipelines using MuJoCo, Isaac Sim, Isaac Lab.
• Collaborate with control, perception and hardware teams for deployment.
• Evaluate experiments and improve capabilities with real-world testing.
• Contribute to scalable manipulation architectures for dynamic environments.

Skills

Python

C++

PyTorch

TensorFlow

ROS/ROS2

Reinforcement Learning

Kinematics & Motion Planning

Perception & Manipulation

Simulation Environments

Vision-Language-Action models

Education

Bachelor’s, Master’s, or PhD in Robotics, Computer Science, Mechatronics, ML

Tools

MuJoCo

Isaac Sim

Isaac Lab

Job description

Role Summary

We are seeking a talented Robotics Manipulation Engineer to develop, implement, and optimise advanced AI algorithms that enable humanoid robots to perceive, plan, and execute complex manipulation tasks. The Robotics Manipulation Engineer will work with cutting-edge techniques, including reinforcement learning (RL), diffusion and flow models, and vision-language-action (VLA) models.

This role bridges AI research and real-world robotic deployment, focusing on developing robust, scalable manipulation behaviours for next-generation humanoid systems.

Key Responsibilities

• Develop manipulation algorithms for high-degree-of-freedom humanoid hands, including grasping, pick-and-place, and bimanual manipulation tasks.
• As a Robotics Manipulation Engineer, build, train, and optimise reinforcement learning policies and generative models for robust manipulation.
• Integrate vision-language-action models into perception-to-action pipelines to enable instruction-driven robotic manipulation.
• Develop and maintain sim-to-real pipelines using simulation environments such as MuJoCo, Isaac Sim, and Isaac Lab.
• Collaborate with control, perception, and hardware engineering teams to deploy manipulation algorithms on physical robotic platforms.
• Evaluate system performance, design experiments, analyse results, and iteratively improve manipulation capabilities through real-world testing.
• Contribute to the development of scalable manipulation architectures for complex, dynamic environments.

Essential Qualifications

• Bachelor’s, Master’s, or PhD in Robotics, Computer Science, Mechatronics, Machine Learning, or a related technical discipline.
• At least two years of relevant experience in robotics, AI, or robotic manipulation.
• Strong programming skills in Python and/or C++, with experience using machine learning frameworks such as PyTorch or TensorFlow.
• Practical experience training reinforcement learning policies, designing reward functions, and working with simulation environments.
• Knowledge of ROS/ROS2 and experience deploying algorithms on real robotic systems.
• Strong understanding of robot kinematics, motion planning, perception, and manipulation control.
• Ability to translate research concepts into reliable, real-world robotic capabilities.

Preferred Qualifications

• Experience working with vision-language-action models, foundation models, or large-scale machine learning approaches for robotics.
• Research publications in relevant robotics, machine learning, or AI conferences.
• Familiarity with real-world robotic experimentation, hardware integration, and sim-to-real transfer.
• Experience developing manipulation capabilities for humanoid robots or other high-degree-of-freedom robotic platforms.

Ideal Candidate

The ideal Robotics Manipulation Engineer combines strong robotics fundamentals with hands-on experience in modern AI techniques. They will be comfortable working across research, simulation, and hardware deployment, developing manipulation systems that perform reliably outside controlled laboratory environments.

This Robotics Manipulation Engineer opportunity is suited to someone motivated by challenging problems in embodied AI, dexterous manipulation, and the development of increasingly capable autonomous robotic systems.