AI Robotics Data Engineer - Multimodal Preprocessing

Persona AI Inc; Houston (TX); On-site; USD 120,000 - 180,000; Full time

Benefits offered by this job

Performance-based bonus

Medical benefits 99% covered

Equity

Competitive PTO

Winter break (Dec 24–Jan 2)

Job summary

Persona AI is building humanoid robots for heavy industry environments, handling welding, maintenance, inspection, and material handling. We seek an AI Data Engineer to turn raw multimodal data into training-grade assets for our robots, enabling faster learning of dexterity and robust perception across teleoperation data.

The role focuses on designing preprocessing infrastructure, validating multi-sensor data, and scaling datasets with advanced augmentation and cross-modal alignment.

Qualifications

• MS or PhD in CS/ML/Robotics or related field.
• Proficient in Python and PyTorch with multimodal data loaders.
• Experience with force-time-series data from sensors and alignment with visual data.
• Experience with OpenCV, FFmpeg, and high-bandwidth video pipelines.

Responsibilities

• Design cross-modal validation systems to verify video, proprioception, force/haptic signals, and language annotations align.
• Coordinate hand-tracking, depth estimation, 3D reconstruction, and pose-tracking modules for robot trajectories.
• Implement data augmentation strategies to expand expert trajectories and improve model robustness.
• Build unified state–action representations for teleoperation data across embodiments.
• Create tooling to query, visualize, and audit datasets and drive re-collection requests.
• Architect end-to-end data ingestion pipelines producing training-ready datasets.

Skills

Python

PyTorch

Time-series processing

Video processing

3D geometry / robotics data

Data augmentation

Multimodal pipelines

Education

MS or PhD in CS/ML/Robotics or related field

Tools

OpenCV

FFmpeg

Decord

Open X-Embodiment / EgoDex

Job description

Persona AI is building humanoid robots for heavy industry environments, handling welding, maintenance, inspection, and material handling. We seek an AI Data Engineer to turn raw multimodal data into training-grade assets for our robots, enabling faster learning of dexterity and robust perception across teleoperation data.

The role focuses on designing preprocessing infrastructure, validating multi-sensor data, and scaling datasets with advanced augmentation and cross-modal alignment.