Data Annotation Specialist

The role

You will iterate on the AI system by annotating data on various diverse tasks performed by robots. You will work with engineering and research teams to ensure data is labeled to required standards.

What you will do

• Manually annotate video sequences including boxes, masks, keypoints, track IDs, and actions and temporal segments.
• Maintain data integrity by applying guidelines and QC checks.
• Resolve ambiguities and fix errors.
• Leverage pre-annotation and autolabeling tools to boost throughput.
• Validate and correct model prelabels.
• Tune auto-tracking and segmentation pipelines.

Required qualifications

• Associate's or Bachelor's degree (or equivalent experience)
• Strong attention to detail
• Consistent application of guidelines
• Ability to follow detailed instructions and work independently with minimal supervision
• Clear written communication
• Collaborative attitude

Nice to have

• Video Annotation Experience: Hands-on experience annotating video (boxes/masks/keypoints, action labels, ID tracking)
• Proficiency with annotation tools; comfort with pre-annotation/autolabel review and correction
• Familiarity with QA practices (inter-annotator agreement, spot checks, golden sets)
• Knowledge of common annotation formats (e.g., COCO, YOLO, MOT/KITTI) and basic video concepts (frame rate, codecs)

About the company

Dyna Robotics builds general-purpose robots powered by a proprietary embodied AI foundation model. The team comes from Google DeepMind, Meta, and Cruise.