The work
Mercor is staffing this project with a leading AI lab that is training and evaluating generative music models. Your day-to-day is close listening and close reading: you receive AI-generated lyrics and audio, often alongside the prompt that produced them, and score them against a written rubric. Typical tasks include rating lyrical craft and creativity, judging whether the output actually did what the prompt asked, and flagging outputs that read as translated-sounding, awkward, or generically "Eurovision Hungarian" rather than something a working Hungarian songwriter would write.
A recurring task type is similarity checking — comparing generated lyrics against published Hungarian songs to identify borrowed lines, structures, or hooks. This requires actual recall of the repertoire, not just search skills. Another is naturalness assessment at the line level: word choice, stress and syllable fit, slang, regional expressions, register shifts, and whether rhymes land in spoken Hungarian rather than only on the page. Rubrics are written in English; your judgments and written justifications will involve both languages.
What the screen looks for
- Verifiable music credentials. Songwriting, lyric writing, performing, composing, or music journalism — with specifics: releases, credits, publications, venues, or bylines.
- Native-level Hungarian. There is a Bilingual Competency Interview conducted in Hungarian; it probes idiom, register, and your ability to explain linguistic judgments precisely.
- Scene knowledge. Contemporary Hungarian genres and sub-genres, current and canonical artists, and how lyrical conventions differ across them.
- Rubric discipline. Whether you can apply someone else's scoring criteria consistently instead of substituting your own taste, and whether you can write a short, defensible justification for a score.
Logistics
Fully remote and asynchronous. Start is immediate; the project runs up to six months. Commitment is flexible — most experts work around 20 hours per week and there is no upper cap. You need headphones or studio monitors suitable for critical listening. Onboarding details typically follow within a few days of completing the application steps. Pay of $20–36/hr reflects observed rates on this platform and is not guaranteed for any individual.