micro1 recruits marketing, sales, and corporate strategy practitioners to write and grade the kind of commercial reasoning tasks AI models still get wrong — positioning briefs, pipeline diagnostics, pricing calls, go-to-market plans. Work is project-based and remote, with observed rates of $90–140/hr depending on seniority and how much of the work is authoring versus review.

Summary, eligibility guidance and screening preparation are prepared by Sidequest from the platform's posting. Check the source listing before applying.

What the work actually involves

You are building and scoring the commercial half of a model's judgment. A typical batch might ask you to write a realistic prompt — a Series B SaaS company with flat net revenue retention, a CFO asking whether to cut the channel program, a category-creation launch with no comparable benchmarks — then produce a reference answer that a partner at a strategy firm or a VP of marketing would sign off on. Other batches run the other direction: you read two model responses to someone else's prompt and rank them, with written justification that survives a second reviewer's audit.

The hard part is not writing marketing copy. Models write fluent copy. What they do badly is choose the right frame, refuse a question that lacks the data to answer it, notice that the stated growth target is arithmetically impossible given the pipeline coverage, or distinguish a positioning problem from a pricing problem. Your grading rubrics get pushed against exactly those failures, which means your reasoning needs to be legible — a score with no explanation of the mechanism is not usable training signal.

What the screen looks for

Operator experience, not commentary. Owned a number, ran a budget, sat in the pipeline review. Consulting, in-house growth, RevOps, brand, corp dev all qualify; content-about-marketing generally does not.
Specificity under follow-up. The AI interviewer will ask what the number was, what you decided, and what happened next. Vague ownership collapses quickly.
Calibration. Willingness to say a task is underspecified, or that two answers are genuinely equivalent, rather than manufacturing a differentiator.
Written English that holds up. Rationales are the deliverable and they are read by strangers.

Logistics

Fully remote and almost entirely asynchronous. Onboarding is a short screening interview conducted by micro1's AI interviewer, then a paid or unpaid calibration set before live batches open. Most contributors work 10–20 hours a week against batch deadlines rather than fixed shifts; volume is genuinely uneven, and periods with no available work happen. Pay is hourly at the observed band, invoiced per project — treat it as contract work, not a retainer.

Marketing, Sales & Strategy Specialist

What the work actually involves

What the screen looks for

Logistics

Interview prep — 10 questions