Machine Learning Engineer - Robot Manipulation

Responsibilities

• Design and implement machine learning algorithms, focusing on reinforcement learning (RL) and imitation learning (IL), for robotic manipulators in dynamic environments.
• Translate high-level objectives into ML problems and deploy robust, scalable models to real-world robotic systems.
• Integrate ML solutions into robotics workflows, ensuring performance in simulated and real-world settings.
• Drive innovation by applying latest ML research to robotic manipulation.
• Own critical ML projects from conception to deployment.
• Collaborate across disciplines and mentor junior engineers.

Requirements

Must-have:

• MS or PhD in machine learning, computer science, robotics, or related field.
• Strong experience training and deploying ML models for real-world applications.
• Deep understanding of RL and IL in robotics.
• Proficiency in Python, PyTorch.
• Experience with data collection, preprocessing, and management for ML training.
• Self-starter with problem identification, prioritization, and execution skills.
• Enthusiasm for fast-paced startup environment.

Nice-to-have:

• Familiarity with Gazebo, MuJoCo, sim-to-real transfer.
• Designing reward functions for manipulation tasks.
• Models for noisy, incomplete, or sparse data.
• Deployment to edge devices for real-time inference.
• Accelerating training with GPU, TPU, or accelerators.
• Stable Baselines, RLlib.
• Robotics principles: kinematics, dynamics, control.
• Publications in ML/robotics/RL.