Robotics ML Expert — MuJoCo Simulation Environments

Alignerr is seeking a specialist who can blend robotics expertise with machine‑learning know‑how to craft sophisticated MuJoCo‑based virtual worlds that teach intelligent agents to move, grasp and cooperate. The position is a fully remote, hourly contract that can range from ten to forty hours each week, allowing the contributor to work from any location while supporting cutting‑edge research in embodied AI. Each day will involve constructing and refining virtual testbeds for robot research, integrating and adjusting reinforcement‑learning methods such as PPO, SAC or TD3, and shaping the reward, observation and action definitions that yield policies capable of crossing from simulation to reality. The role also requires troubleshooting contact dynamics, fine‑tuning actuator models, and assessing the robustness and generalisation of trained behaviours. Clear written records of environment layouts, training pipelines and experimental outcomes are essential, as is asynchronous