Remote Vision Data Annotator for AI Training

Insight Global; United States; Remote; USD 28,000 - 34,000; Full time

Job summary

Insight Global is seeking Vision Data Specialist to support AI data labeling for computer vision models. The role focuses on high-volume image and video annotation, including bounding boxes, segmentation, classification, object tracking, and QA/review according to strict guidelines.

The ideal candidate is detail-oriented, able to work remotely from the United States, and comfortable aligning to PST hours while delivering accurate, scalable results in a fast-paced environment.

Qualifications

• Bachelor's degree; open major allowed.
• 2
• 6+ months of computer vision, image/video annotation, GIS labeling, mapping, QA/QC, or data annotation.
• Ability to follow detailed guidelines and make consistent labeling decisions across bounding boxes, classification, tracking, and QA.

Responsibilities

• Accurately annotate CV datasets using bounding boxes, segmentation, classification, and tracking.
• Review content and apply project-specific guidelines.
• Identify edge cases and escalate ambiguities.
• Maintain high quality, throughput, and compliance with metrics.
• Collaborate with peers to improve annotation workflows and provide tooling feedback.
• Complete assignments within service-level expectations and project deadlines.

Skills

Image Annotation

Video Annotation

Bounding Boxes

Object Classification

Object Tracking

QA/QC

Attention to detail

Education

Bachelor's Degree

Tools

Annotation Tools

Job description

Insight Global is seeking Vision Data Specialist to support AI data labeling for computer vision models. The role focuses on high-volume image and video annotation, including bounding boxes, segmentation, classification, object tracking, and QA/review according to strict guidelines.

The ideal candidate is detail-oriented, able to work remotely from the United States, and comfortable aligning to PST hours while delivering accurate, scalable results in a fast-paced environment.