Data Annotator

Insight Global; United States; On-site; USD 25,000 - 39,000; Full time

Job summary

Insight Global seeks Data Annotators to support AI and computer vision model development by labeling and validating video and text datasets used to train systems.

The role requires a detail-oriented mindset, the ability to work in a fast-paced setting, and consistent accuracy when handling large-scale visual data. A Bachelor's degree and 3 years of annotation experience are preferred.

Qualifications

• Bachelor's degree required.
• 3 years of data annotation experience.
• Working knowledge of Python or SQL or a Computer Science background.

Responsibilities

• Review model responses, research evidence, images, videos, or agent actions according to the assigned rubric.
• Apply labels, rankings, scores, or timestamps and provide concise explanations supported by the available evidence.
• Flag unclear instructions, uncertain cases, and recurring errors to the task lead. Apply approved guidance and complete targeted rework.
• Participate in training and calibration, incorporate feedback, meet agreed deadlines, and communicate blockers early.

Skills

Detail oriented

Fast-paced

Python/SQL awareness

Education

Bachelor's degree

Tools

Python

SQL

Job description

We are seeking Data Annotators to support the development of next-generation AI and computer vision models. This role will focus on labeling and validating video and reason based text datasets used to train machine learning systems.

The ideal candidate is highly detail-oriented, thrives in fast-paced environments, and can consistently make accurate decisions when working with large-scale visual datasets.

Must-Have Qualifications

• Bachelor's Degree required.
• 3 years of data annotation experience
• Working Knowledge of Python or SQL or a Computer Science Background

What annotators will do

Review model responses, research evidence, images, videos, or agent actions according to the assigned rubric.

Apply labels, rankings, scores, or timestamps and provide concise explanations supported by the available evidence.

Flag unclear instructions, uncertain cases, and recurring errors to the task lead. Apply approved guidance and complete targeted rework.

Participate in training and calibration, incorporate feedback, meet agreed deadlines, and communicate blockers early.