Data Annotators: Paid Tasks for AI Model Training

What you'll do

You log into Terac's remote platform and work through a queue of annotation assignments. What is in the queue changes by batch: the ad lists categorising text, highlighting specific entities and evaluating AI-generated responses for quality as examples. Guidelines evolve as the project does, so a large part of the job is re-reading instructions and applying them consistently. The ad says the work is meant to improve model accuracy, safety and reasoning, and does not name the customer.

Who fits

• Prior experience in data annotation, data labeling or AI model evaluation
• Freelance labelers, QA testers, content moderators and ML data specialists are named as welcome
• Comfort working alone and adapting when labeling rules change

What it pays

$40 per hour, a single figure. The ad does not say whether hours are tracked or whether it is a rate converted from task pay, nor how many hours are available. Rolling availability suggests work comes in as batches open rather than as a fixed schedule.

Worth knowing

Good:

• A clear hourly figure, higher than many general annotation gigs
• Self-paced; fits around other work
• Ongoing project with rolling availability
• No country restriction

Less good:

• Prior annotation or evaluation experience is expected, so it is not an entry route
• The ad gives no hours, volume or contract length
• Changing guidelines mean quality standards move under you
• Valid through 24 November 2026

About this listing

Terac posted this on 25 September 2026; we read it on 23 October. The Apply button opens this study on Terac; you sign in or create a Terac account there to apply. Registering does not guarantee a match. See Terac.