What the work actually involves
You will be handed artifacts of the kind a small-business operator produces or receives: a three-statement model in Excel, an investor update, a supplier agreement, a hiring plan, a customer acquisition analysis in PowerPoint. Sometimes the artifact was written by a person and sometimes by a model, and you generally will not be told which. Your job is to apply a rubric — accuracy, completeness, whether the reasoning survives contact with how businesses actually work — and to write feedback that a non-expert reviewer can follow. A common task shape is comparing two candidate responses to the same prompt and explaining, in specific terms, why one is better. Another is authoring a realistic prompt from your own operating history and then writing the reference answer against it.
The distinguishing skill is not eloquence. It is catching the plausible-but-wrong: a cash flow projection that ignores payment terms, a term sheet summary that misreads a liquidation preference, a go-to-market plan that assumes a sales cycle no one in that industry has. Vague praise and vague criticism are equally useless to the people training on your output.
What the platform screens for
Mercor's screen is AI-led and largely conversational, and it presses on specifics. Expect to be asked what you actually founded, what it did, how long it ran, what revenue looked like, and what you personally handled versus delegated. Follow-ups go one layer deeper than most candidates prepare for — if you say you managed cash flow, expect a question about your collections process or your worst month. Your resume and any linked profiles are cross-checked against what you say. Solo consultancies, e-commerce stores, agencies, and franchises count; "co-founder" of something that never had a customer or a bank account is a hard sell.
Logistics
- Fully remote, asynchronous, contractor engagement — no set schedule, but tasks carry deadlines
- Volume fluctuates by project; many contributors report 10–20 hours a week when a project is live, with gaps between
- Observed rates run $80–160/hr, set per project and per contributor, not guaranteed
- Eligibility is limited to people based in the US, Europe, UK, or Canada
- Onboarding includes calibration tasks; sustained agreement with reviewer consensus is what keeps you on a project