Staff Robotics ML/Data Engineer — Multimodal Data Architect

Persona AI, Inc.; Houston (TX); On-site; USD 150,000 - 210,000; Full time

Benefits offered by this job

Competitive compensation

Performance-based bonus

99% employer-covered medical benefits

Equity

PTO & winter break

Job summary

Persona AI, Inc. is seeking a Staff, Robotics ML/Data Engineer to shape data-intensive training pipelines for humanoid robots.

Based in Houston or Pensacola, you will architect multimodal data ingestion, validation, and augmentation to accelerate model learning across teleoperation and manipulation tasks. Join a team building robotic platforms for heavy industry with equity, strong benefits, and winter break.

Qualifications

• MS or PhD in CS/Robotics/ML or related field.
• Strong Python skills and PyTorch with multimodal dataloaders.
• Experience processing force/torque and time-series data from sensors.
• Expertise in video processing with OpenCV/FFmpeg/Decord.
• Strong knowledge of 3D geometry, transforms, URDF.
• Experience implementing CV/time-series data augmentation.

Responsibilities

• Design cross-modal validation systems ensuring video, proprioception, force/haptic signals, and language annotations agree with each other.
• Retarget and align human demonstrations to robot trajectories with kinodynamic validation.
• Implement robust data augmentation to expand expert trajectories and improve model robustness.
• Build teleoperation synchronization and unified state–action representations across different embodiments.
• Create tooling to query, visualize, and audit datasets and drive targeted re-collection requests.
• Architect end-to-end multimodal data pipelines from raw recordings to training-ready datasets.

Skills

Python

PyTorch

Multimodal Dataloaders

Time-series processing

Video processing

OpenCV/FFmpeg/Decord

3D geometry

Data augmentation

Education

MS or PhD in CS/Robotics/ML

Tools

Open X-Embodiment

DROID

AgiBot World

EgoDex

Omniverse

MuJoCo

Ray

Job description

Persona AI, Inc. is seeking a Staff, Robotics ML/Data Engineer to shape data-intensive training pipelines for humanoid robots.

Based in Houston or Pensacola, you will architect multimodal data ingestion, validation, and augmentation to accelerate model learning across teleoperation and manipulation tasks. Join a team building robotic platforms for heavy industry with equity, strong benefits, and winter break.