What the work involves

You receive artifacts an AI produced in response to a nonprofit or philanthropic prompt: a letter of inquiry to a family foundation, a theory-of-change diagram, a program evaluation summary, a multi-year operating budget, a slide deck for a board finance committee. Your job is to score each against a rubric and write up what's wrong. The interesting failures are rarely typos — they're logic models where outputs are labeled as outcomes, indirect cost rates that no funder would accept, restricted-fund language applied to unrestricted gifts, IRS Form 990 references that don't match the filing, or a deck that's factually fine but formatted in a way no program officer would take seriously.

Expect a mix of grading and open-ended critique. Some tasks ask for a numeric score plus a paragraph; others ask you to rewrite a section correctly so the model has a reference. Because a large share of artifacts are slide decks, comfort with Google Slides and PowerPoint mechanics — master layouts, chart formatting, alignment, readable density — matters as much as subject knowledge.

What the platform screens for

  • Verifiable experience. Mercor's screening walks through your actual work history: which organizations, what budget size, whether you were on the grantseeking side, the grantmaking side, or both.
  • Depth under follow-up. Expect the interviewer to push a level past your first answer — if you mention outcome measurement, be ready to describe an indicator you personally defined and how it held up.
  • Evaluation judgment. Can you separate "wrong" from "stylistically not how I'd do it," and can you rank errors by severity rather than listing them flat?
  • Presentation fluency. Concrete evidence you build decks and financial spreadsheets yourself, not just review them.

Logistics

Remote, hourly, asynchronous. Observed pay for this band has been $80–120/hr, set by Mercor based on credentials and screening outcome — not guaranteed, and not something Sidequest sets. Work arrives in batches; most evaluators pick up 5–20 hours a week with no fixed schedule, though task supply fluctuates by project. Some projects require sustained availability over a several-week window, which the screen will ask about directly.