Robotics ML Expert (MuJoCo Environments) | Remote Contract

Project Overview We are looking for highly experienced Robotics Machine Learning Experts to work on advanced AI simulation and reinforcement learning projects focused on MuJoCo environments. This is a high-level remote contract opportunity where you will help design, build, and optimize physics-based simulation environments used to train intelligent agents. The work directly contributes to cutting-edge research in robotics, embodied AI, and sim-to-real learning systems. You will be working on tasks involving locomotion, manipulation, multi-agent systems, and reinforcement learning pipelines used to train next-generation AI systems. Compensation $100 – $150 per hour Remote contract role Flexible working hours 10–40 hours per week Long-term extension opportunities based on performance Responsibilities Design and develop MuJoCo simulation environments for robotics AI training Implement and optimize reinforcement learning algorithms (PPO, SAC, TD3, etc.) Define and tune reward functions, observation spaces, and action spaces Debug physics simulations including contact dynamics and actuator behavior Work with MJCF/XML model files and environment configurations Evaluate trained policies for robustness and sim-to-real transfer potential Document experiments, environment design, and training procedures clearly Collaborate with remote research and engineering teams Stay updated with advancements in robotics, RL, and embodied AI Required Skills Strong hands-on experience with MuJoCo or similar simulators (dm_control, Gymnasium Robotics) Deep understanding of Reinforcement Learning (RL) algorithms Strong Python programming skills Experience with PyTorch or JAX Knowledge of robot kinematics, dynamics, and control systems Experience designing reward functions for complex tasks Ability to work with MJCF/XML simulation files Strong analytical and debugging skills Ability to work independently in a remote environment Strong technical communication and documentation skills Preferred Experience Sim-to-real transfer techniques (domain randomization, system identification) Experience with Isaac Gym, PyBullet, Drake, or Genesis Multi-agent reinforcement learning Imitation learning or model-based RL Research publications in robotics or machine learning Open-source contributions in RL or robotics frameworks Graduate-level education in Robotics, ML, or Computer Science Ideal Candidate You are a strong fit if you: Have hands-on experience building RL environments Understand how physics simulation impacts learning performance Can design reward systems that produce stable and transferable policies Are comfortable working independently in research-style environments Enjoy solving complex problems in robotics and AI systems Why Join This Role? Work on cutting-edge robotics and AI research problems Fully remote and flexible contract structure High-impact work influencing real-world AI systems Collaborate with top-tier global ML and robotics practitioners Opportunity for long-term contract extensions and new research projects Work at the intersection of simulation, robotics, and reinforcement learning Application Instructions To apply, please include: Brief introduction and relevant experience Details of MuJoCo or simulation-based projects you’ve worked on Experience with RL algorithms and frameworks Python and ML framework expertise (PyTorch/JAX) Any research papers, GitHub, or portfolio links (if available) LinkedIn profile (optional) Applications are reviewed on a rolling basis. Qualified candidates may be contacted for further technical evaluation.

Project ID: 40465280