Software Engineer, Robotics Data
Core
Build the data backbone for physical AI by creating pipelines that ingest raw multi-sensor capture (video, depth, inertial, audio) from the field and deliver validated, privacy-safe datasets to frontier labs.
Role type
Senior backend/data engineer specializing in physical-world sensor data pipelines
Builds
End-to-end sensor data pipelines, automated QC systems, dataset schemas, and shared processing components for petabyte-scale data
Domain
AI data infrastructure, robotics, physical-world sensing, multi-sensor fusion
Deliverable
production ML models
Required skills
High-volume data pipeline ownership, video/sensor data processing at scale, Python, AWS, distributed batch processing, object storage economics, automated quality control design, data schema/versioning/provenance management, privacy redaction, format packaging
Preferred skills
Robotics data formats (MCAP, ROS bags, protobuf, Foxglove), camera geometry, multi-sensor calibration and synchronization, computer vision, multimodal ML, VLM-based labeling or QC, data engines for AV/robotics/egocentric video
Technologies
Python, AWS, MCAP, ROS bags, protobuf, Foxglove
Responsibilities
Ingest raw multi-sensor capture from field devices, design automated QC for timing/sync/calibration health, establish dataset schemas and versioning, build shared processing components for privacy redaction and encoding, integrate VLM-assisted pre-labeling into production workflows
Seniority
Senior, hands-on IC