Deep Learning - Robot Manipulation Engineer

2 months ago

Houston, TX, USAMid Level

H1B Sponsor

Responsibilities

Design and implement deep learning models and training procedures for dexterous manipulation by high-degree-of-freedom humanoid robots with multi-fingered hands.
Train models for object interactions, precise grasping, long-horizon tasks, tool use, and in-hand object manipulation using curriculum learning.
Integrate tactile sensing and proprioception into end-to-end learning pipelines for robust closed-loop policies.
Develop data collection and versioning strategies with teleoperation and data teams.
Leverage existing manipulation models and develop new architectures for complex tasks.
Deploy trained models to robotic hardware and integrate them with control systems and sensors for real-time, safe operation.
Optimize the manipulation machine learning pipeline.
Build evaluation pipelines using simulation and real-world trials to measure performance, robustness, and generalization.
Help attract, nurture, and grow machine learning and autonomy teams.

Master’s or PhD in robotics, computer science, or a related field.
At least 3 years of experience applying deep learning to robotic manipulation.
Strong understanding of behavior cloning, vision-language-action models, diffusion policies, foundation models, and related state-of-the-art algorithms.
Experience with cloud computing and large-scale datasets.
Understanding of deploying neural network models in real-world environments.
Ability to write high-quality software.
Experience with computer vision, sensors, point clouds, segmentation, and object detection is a bonus.
Published papers at ICML, ICRA, CoRL, RSS, NeurIPS, or other leading ML and robotics conferences are a bonus.
Experience deploying robots, collecting large datasets, and training production neural networks is a bonus.

Competitive compensation and performance-based bonus.
99% employer-covered medical benefits, early-stage equity, and competitive paid time off.
Company-wide paid winter break from December 24 through January 2.
Full-time role based in Houston, Texas, with 10% travel.
Opportunity to shape deep learning and manipulation technology for humanoid robots.