Education — Visual Document Understanding
Mercor
$60/hour · as listed
Help train AI models to understand visual documents in educational contexts. This intermediate-level data work involves creating benchmark datasets and evaluation tasks—you'll design complex scenarios with clear correct answers and scoring rubrics that test how well AI systems can read and follow instructions on images of educational materials. The role suits people with coding background who enjoy structured problem-solving and attention to detail.
This is remote, part-time work available to US-based professionals, typically 15–20 hours per week, listed at $60/hour. You'll work on Mercor's platform as part of a team evaluating AI performance in visual document understanding, contributing to datasets that help improve these models' real-world capabilities.
From the listing: Mercor marketplace · part-time · 20h/week. Benchmark dataset project evaluating AI models on visual document understanding and instruction-following in the Education domain. Experts author complex, grounded tasks with a clear ground-truth output and objective rubric. ~15–20 hrs/week, remote, US/Canada.
Applications are completed on the listing’s own site. Pay is shown as posted by the source (“as listed”) or the aggregator’s estimate where marked — offers and availability change, and individual results vary.
Want help getting selected — and finding the best-paid work for your country? See how membership works →