Member of Technical Staff, Robotics Engineer
New York, NY • FullTime
Posted 3mo ago
Remote Work Policy
On-site
Employment Type
FullTime
Categories
Applied AI Engineer
About the job
We are seeking a Robotics Engineer to bridge the gap between our video-native foundation models and real-world hardware. This role involves deploying world-model-based policies on actual robots, ensuring their functionality in physical environments. You will engage with the entire spectrum of robot learning, from data collection and task design to policy deployment and physical evaluation. This is a hands-on, execution-focused position at the confluence of foundation models, learned robot policies, and hardware, requiring deep robotics expertise to deliver end-to-end robot policy solutions for applications like manipulation and mobile robotics.
Responsibilities
- Own the deployment loop for learned policies on real robot arms, including inference serving, action decoding, controller integration, latency management, and safety.
- Diagnose and resolve real-world policy failures by identifying issues in data coverage, sensor calibration, or execution, and collaborating with the research team for fixes.
- Design demonstration data collection protocols, including task setup, camera positioning, and teleoperation conventions, to generate effective training data for world models.
- Collaborate with research to conduct controlled experiments linking world model representations, data composition, and fine-tuning to physical success rates.
- Define and execute physical evaluation protocols to ensure trustworthy and comparable results across policy iterations.
- Adapt robot policy frameworks to work with partner hardware and new robotic embodiments.
Requirements
- Proven experience deploying software on real robot hardware and iterating to achieve functionality, including learned policies, motion planning, classical control, or perception-driven manipulation.
- Proficiency in the software-to-robot interface, understanding action/command spaces, control frequencies, observation pipelines, and calibration effects on real-world behavior.
- Comfort working with both software and physical systems, capable of reconfiguring robot workspaces and troubleshooting policy failures.
- Experience deploying learned policies (VLAs, diffusion policies) on real hardware is a plus.
- Experience with video/multimodal generative models or world models is a plus.