What the work involves

You will be handed advanced economics prompts — the kind a strong graduate student or working researcher would struggle with — and asked to produce a complete, defensible answer. These become "golden responses": the benchmark output a model is trained toward. Expect to move between subfields: a comparative statics derivation one day, an identification critique of a difference-in-differences design the next, then a question on monetary transmission or bargaining models. Alongside authoring, you'll review other contributors' work and model outputs, flagging errors in reasoning, misapplied assumptions, stale methodology, or answers that are technically correct but pedagogically useless.

A large share of the value is in the write-up. You're expected to document why you answered as you did — which assumptions you invoked, where the literature disagrees, what a common wrong answer looks like and why it fails. Annotation leads and PMs will iterate with you on spec, and there are occasional remote workshops to align contributors on standards. micro1 states explicitly that no prior AI or annotation experience is required.

What the screening looks for

  • Verified credential. PhD in economics or a very close adjacent field; expect to name your institution, advisor, and dissertation topic.
  • Depth under follow-up. The AI interviewer will push past your first answer — vague or textbook-recited responses are the main failure mode.
  • Breadth across subfields. Comfort in at least two or three of micro theory, macro, econometrics, behavioral, labor, IO, or international.
  • Written precision. Evidence you've produced peer-reviewed papers, research reports, or teaching materials that others relied on.
  • Realistic availability. Minimum weekly task submission applies, and onboarding-to-first-task is 24–48 hours.

Logistics

Fully remote, contractor status, asynchronous. Compensation is per accepted task rather than hourly; the $80–90/hr band reflects observed effective rates for experienced contributors and is not guaranteed — slower workflows or heavy revision cycles pull it down. There is a weekly minimum submission requirement, so this suits people who can reliably carve out several focused hours a week rather than sporadic bursts. Roles are typically filled within 48 hours of application.