What the work actually involves

You install a specialized build of an AI assistant app on your own Mac and use it the way a real person would — dictating emails and messages, asking it to summarize a document, generate an image, reformat text, or apply a spreadsheet formula. Everything is done by voice, in Italian. The point is to produce natural, unscripted speech rather than clean test sentences, so the assistant is exposed to how Italians actually talk to their devices: elisions, dialect-tinged phrasing, self-corrections, background noise.

The second half of the job is the documentation, and it is where most of the hours go. For each task you record the audio, write a verbatim transcript of what you actually said, produce a cleaned-up version, rate the assistant's response, and log the whole thing in a structured tracking sheet with recordings uploaded to a shared drive. Rows that are mislabelled, missing a recording, or transcribed loosely are not usable data. Self-directed quality review is explicitly part of the role — nobody is checking your sheet row by row before it ships.

What the screening looks for

Mercor runs a resume upload plus a roughly 25-minute conversational AI interview covering background, experience and motivation. For this listing the screen is weighted toward three things: genuine native Italian from Italy (accent, idiom, regional register all matter), enough English to follow written task instructions and rubrics, and hard confirmation that you own an Apple Silicon Mac — M1, M2 or M3 — running macOS Sequoia 15.0 or newer. The hardware requirement is not negotiable and is checked during technical setup, not just asserted in the interview. Expect follow-ups on how you habitually use productivity apps and on your tolerance for repetitive structured data entry.

Logistics

  • Fully remote, self-scheduled, on your own machine
  • Fixed four-week duration with an immediate start
  • Hourly, observed at $21/hr; hours depend on task allocation
  • Onboarding includes a Mercor Expert email and a technical setup step; next steps typically arrive within a few days
  • You need a reasonably quiet space to record voice tasks and a stable connection for uploading audio