Machine Learning Engineer

Key Responsibilities

• Develop state-of-the-art reinforcement learning algorithms for robotic control, planning, and decision-making.
• Build and maintain simulation environments (e.g., Isaac Gym, Mujoco, Gazebo, PyBullet) for scalable RL training.
• Train RL agents using deep learning frameworks (PyTorch / TensorFlow) and optimize them for sim-to-real transfer.
• Collaborate with robotics and controls engineers to integrate RL policies on real robotic hardware (arm manipulators, humanoids, etc.)
• Implement robust RL pipelines including data collection, reward shaping, distributed training, and evaluation.
• Perform benchmarking, diagnostics, and troubleshooting of RL algorithms under various operational conditions.
• Research and apply methods such as offline RL, hierarchical RL, imitation learning, model-based RL, or multi-agent RL.
• Optimize for safety, performance, and real-time constraints in robotic applications.
• Document technical findings, results, and best practices.

Required Qualifications

• Bachelor’s or Master’s degree in Computer Science, Robotics, Machine Learning, Electrical Engineering, or related field.
• Strong experience with Reinforcement Learning and deep learning (DDPG, PPO, SAC, TD3, A3C, etc.).
• Proficiency in Python and RL/ML frameworks (PyTorch preferred).
• Hands-on experience with robotics simulation tools (Isaac Gym, Mujoco, PyBullet, Gazebo, or others).
• Solid understanding of kinematics, control theory, or robot dynamics.
• Experience integrating algorithms with real robotic hardware (ROS/ROS2 experience strongly preferred).
• Strong debugging skills and ability to analyze complex system behaviors.