Remote Data Annotation Analyst — 2D/3D Vision for AI

Jobgether; Lavamünd; Remote; EUR 18.000 - 30.000; Vollzeit

Benefits dieser Stelle

Remote work in US

Company-provisioned machine

Access to AI training projects

Zusammenfassung

Partner Company seeks a Data Annotation Analyst to create high-quality ground-truth datasets for training perception models using camera imagery, video, and LiDAR data. You will label objects, segments, poses, and keypoints with strict adherence to annotation guidelines.

You will collaborate with QA and fellow annotators, participate in calibration sessions, and help improve guidelines while maintaining production quality and consistency.

Qualifikationen

• Strong attention to detail and accuracy in repetitive, visually intensive work.
• Strong spatial reasoning for 3D environments and object relationships.
• Ability to learn specialized annotation software quickly.
• Experience with image annotation or GenAI data workflows.
• Reliable high-speed internet and distraction-free workspace.
• Experience with sensor data like LiDAR is a plus.

Aufgaben

• Annotate 2D and 3D objects across images, video, and LiDAR point clouds.
• Create and maintain segmentation, tracking, pose, and keypoint annotations.
• Apply detailed annotation specifications including occlusion rules.
• Review work, address QA feedback, and meet quality standards.
• Identify ambiguous scenes and gaps in guidelines; document issues.
• Participate in calibration sessions and guideline reviews.
• Share edge cases to improve guidelines and taxonomies.
• Maintain production volume with high accuracy and consistency.

Kenntnisse

Attention to detail

Spatial reasoning

Software navigation

Annotation workflows

QA collaboration

Autonomous systems

AutoCAD familiarity

3D data understanding

Fast-paced environment

Team standups

Ausbildung

Associate or Bachelor's degree

Tools

AutoCAD

Annotation tools

3D annotation tools

Geospatial tools

Jobbeschreibung

Partner Company seeks a Data Annotation Analyst to create high-quality ground-truth datasets for training perception models using camera imagery, video, and LiDAR data. You will label objects, segments, poses, and keypoints with strict adherence to annotation guidelines.

You will collaborate with QA and fellow annotators, participate in calibration sessions, and help improve guidelines while maintaining production quality and consistency.