The work

Most of your time is spent in a listening seat with a rubric open beside you. You'll receive pairs or batches of AI-generated tracks, often against a text prompt, and judge them on musicality, creativity, prompt adherence, vocal quality, and mix/production quality — then write a short justification that a model trainer can act on. A second stream of work is annotation: tagging genre and sub-genre, song structure, instrumentation, vocal type and delivery, and other musical characteristics. A third stream is Norwegian-language review, checking generated lyrics and vocal performances for grammatical correctness, natural phrasing, pronunciation, and fidelity to reference material.

The judgment calls are rarely obvious. Two tracks may both technically satisfy a prompt while one is competent and forgettable and the other is rough but genuinely interesting. You'll need to separate arrangement problems from mix problems, and recognize when a vocal sounds wrong because of the performance rather than the processing. Consistency across hundreds of comparisons matters more than any single verdict.

What the screen looks for

  • Verifiable production experience — two or more years producing, engineering, or mixing, with specific projects, DAWs, and roles you can name. Credited or commercially released work is a plus.
  • Real Norwegian fluency, assessed in a Bilingual Competency Interview conducted in Norwegian, plus the ability to work from detailed English instructions.
  • Breadth across genres, including contemporary Norwegian music — expect follow-ups on sub-genres and artists.
  • Rubric discipline: whether you can apply criteria you didn't write, hold a consistent bar, and explain a rating rather than just assert it.

Logistics

Fully remote and asynchronous, with an immediate start and a project running up to roughly six months. Most experts log around 20 hours per week; there is no stated cap, and hours are self-scheduled. Observed pay for this listing is $42–78/hr, set by experience and assessment outcome rather than guaranteed. You need headphones or monitors suitable for critical listening and a quiet room; onboarding steps typically arrive within a few days of application.