Music Expert — AI Evaluation & Annotation (Generative Music)

About this role

**Mercor is hiring professional musicians, producers, and musicologists** to evaluate and annotate AI-generated music for a frontier generative-music AI lab. You'll work on structured, high-signal tasks that turn subjective musical quality into data the lab can act on: - **Preference labeling** — given two AI-generated tracks for the same prompt, judge which is stronger on specific dimensions (faithfulness to the source, instruction-following, audio fidelity & production, vocal quality, musicality) and write a clear rationale. - **Music captioning** — write precise, schema-consistent descriptions of a track: genre & style, instrumentation, structure with timings, key/tempo/feel, vocals, and production/era. - **Lyric alignment** — mark word-by-word start/end timing (karaoke-style), correct the transcript where words were dropped or looped, and flag pronunciation and repetition artifacts. **You're a fit if you:** - Are a trained musician with strong critical listening across multiple genres and languages - Can hear and articulate whether an output is faithful, well-produced, and on-spec - Are comfortable applying structured rubrics and writing clear, falsifiable justifications - Bonus: production/mixing experience, formal music theory, or multilingual lyric fluency **Details:** Remote · flexible hours · short calibration (Rubric Academy) before staffing · background check + confidentiality agreement required. Genres and languages matched to your strengths.

Skills & domains

• ai-training
• rlhf
• sme
• annotation
• Arts & Design