ML Engineer – Manipulation

This is a fully remote position, open to applicants in United States.

📋 Description

• Create learning-centric manipulation models for comprehensive sensor-driven interactions (e.g., reaching, motion generation, and execution in dynamic settings).

• Develop and sustain manipulation training pipelines: generating datasets from robot logs/teleoperation, action representations, augmentation, and distributed training processes.

• Design evaluation metrics and regression tests to measure manipulation reliability, recovery behavior, and safety in real-world environments.

• Establish sim-to-real workflows for manipulation learning, which includes simulation environments, domain randomization, and testing for failure modes.

• Optimize and refine models for edge deployment; assess latency, memory usage, and stability on designated hardware.

• Collaborate with the AI platform team to integrate policies with control and safety systems, ensuring validation of end-to-end performance on robots.

• Examine field performance, identify key failure modes, and promote iterative enhancements through data collection and focused retraining.

⛳️ Requirements

• Bachelor’s or Master’s degree in Robotics, Computer Science, Electrical Engineering, or a related discipline (PhD is advantageous).

• Over 3 years of experience in applying machine learning to robotics manipulation, visuomotor control, or sequential-to-sequence models.

• Strong expertise in PyTorch and experience in constructing dependable training/evaluation pipelines.

• Solid software engineering skills in Python; capacity to collaborate effectively across ML and robotics teams.

🏝️ Benefits

• Comprehensive healthcare coverage including medical, dental, and vision.

• Opportunities for professional development and continuous learning.

• Flexible work hours and remote work options.

• Engaging team environment with collaborative projects and open communication.