What the work involves

Most of your time goes to structured listening and reading passes over model outputs. You'll rate AI-generated lyrics on craft dimensions — imagery, rhyme and meter, internal consistency, whether the song actually does what the prompt asked — and flag outputs where the Greek reads like translated English rather than something a Greek songwriter would write. A recurring task is comparison work: checking generated lyrics against published songs to surface close paraphrase, borrowed hooks, or structural lifts that a rubric would treat as insufficiently original.

The language judgment is the part that can't be outsourced to a non-native reviewer. You're assessing word choice, register, slang, and regional expression across contemporary Greek genres — whether a line lands as idiomatic, dated, awkwardly formal, or simply wrong. Written ratings usually need a short justification, so the ability to say why something fails in clear English matters as much as the ear itself.

What the platform screens for

  • Verifiable music background. Songwriting credits, released work, performance or composition training, or a body of music journalism. Vague "lifelong music fan" framing does not clear the bar.
  • Greek at native or near-native level, tested directly through Mercor's Bilingual Competency Interview conducted in Greek.
  • Working knowledge of the current Greek scene — artists, sub-genres, what sounds contemporary versus five years stale.
  • Rubric discipline. Screens probe whether you can apply someone else's standard consistently instead of substituting personal taste.

Logistics

Fully remote and asynchronous, with immediate start and a duration of up to six months. Most experts work around 20 hours per week; there's no stated cap, and scheduling is yours to arrange. You'll need headphones or studio monitors adequate for critical listening. Onboarding details typically follow within a few days of completing the application steps.