What the work actually involves

You'll be handed artifacts that look like the ones you produce at work: a response to an angry enterprise customer, a renewal risk summary, a QBR slide built in PowerPoint, a support macro library, a churn cohort analysis in Excel. Your job is to say whether the artifact would survive contact with a real account — and to write down precisely why it wouldn't. That means catching the SLA commitment nobody could honor, the tone that reads fine to a manager and condescending to a customer, the health score that was calculated from the wrong denominator, the escalation path that skips the account team.

Output is structured: scores against a rubric with several dimensions, plus written justification tied to specific lines or cells. Some projects ask for a corrected version alongside the critique. Others are head-to-head comparisons of two model outputs where you pick a winner and defend the call. Reviewers who write "tone is off" get flagged; reviewers who write "the second paragraph attributes the outage to the customer's configuration before RCA is complete, which contradicts the earlier apology and invites a contract dispute" get repeat work.

What the screen looks for

Mercor runs an AI-led interview before any project assignment. It probes whether your experience is real and operational: which tooling you actually sat in (Zendesk, Intercom, Salesforce, Gainsight, Totango, Freshdesk), what your book of business looked like, whether you owned renewals or handled tier-1 volume, how you were measured. Expect follow-ups that go one level deeper than your first answer — vague claims about "improving CSAT" invite a question about baseline, sample size, and what changed. It also tests calibration: can you separate an output that is merely unpolished from one that is factually or contractually wrong, and can you hold a rubric consistently across twenty items rather than drifting toward the middle.

Logistics

  • Fully remote and asynchronous; no fixed shifts, no customer contact
  • Project-based, typically 10–20 hours per week, with volume that fluctuates between engagements
  • Deadlines are per-batch rather than per-day, so evenings and weekends are workable
  • Eligibility is limited to candidates based in the US, Europe, UK, or Canada
  • Pay is stated as observed on the platform, not guaranteed; rates vary by project, rubric complexity, and seniority