Staff Engineer, Deep Learning Manipulation

2 days ago

Houston, TX, USAStaff+

Responsibilities

Design and implement advanced robot learning models and training procedures for dexterous manipulation by humanoid robots with high degrees of freedom and multi-fingered hands.
Own the ML manipulation pipeline and roadmap, combining learning paradigms to achieve industrial-level robustness.
Work with the Teleoperation and Data team on data collection, sensor modularity, and data-mixture strategies.
Use state-of-the-art manipulation models and develop new architectures for complex tasks.
Deploy trained manipulation models to robotic hardware with real-time performance, safety, and integration with control systems and sensors.
Monitor developments in machine learning research and development.
Build and execute evaluation pipelines using real-world trials and simulation to measure performance, robustness, and generalization.
Help attract, nurture, and grow the machine learning team.

5+ years of experience applying deep learning techniques to robotic manipulation in industrial settings.
Master’s or PhD in robotics, computer science, or a related field.
Strong understanding of and proficiency with behavior cloning, reinforcement learning, vision-language-action models, and foundation models.
Experience with cloud computing and large-scale datasets.
Ability to write high-quality software in Python, C++, or Rust.
Experience with robot simulation, locomotion, computer vision, robot hardware, or sensors is a bonus.
Published papers at top ML or robotics conferences such as ICML, ICRA, CoRL, RSS, or NeurIPS are a bonus.
Experience training models and deploying them to robots in production environments is a bonus.