What the work involves

Each deliverable is a self-contained evaluation task. You start from something you actually did on a desk or in a compliance review — a suitability analysis for a client with mismatched risk tolerance, a futures roll under margin pressure, a books-and-records question, a solicited-vs-unsolicited order marking dispute — and rebuild it as a multi-step problem an AI agent must work through. That means sourcing or constructing authentic supporting artifacts: position blotters, account applications, prospectuses, exchange notices, FINRA or CFTC rule text, price and volume data in spreadsheet form.

The rubric is where most of the effort concentrates. Thirty-five or more discrete, checkable items covering market judgment, compliance sensitivity, disclosure completeness, and how the agent communicates with a hypothetical client. Items need to be binary or near-binary — a reviewer who is not a licensed rep should be able to apply them consistently. You will also write the rationale for the correct action, with regulatory grounding, because that is what makes the task defensible when a model disagrees with you.

What the platform screens for

  • Licensure and desk time. Active Series 7/63 or a non-US equivalent, and roughly four or more years inside a brokerage, bank, or trading platform where compliance workflow was part of the job rather than an adjacent department.
  • Specificity under follow-up. Screens push on details: which supervisory system flagged what, how you documented a suitability call, what you did when a client pushed back. Generic answers about "client relationships" do not survive.
  • Rubric instinct. Whether you can decompose a judgment call into observable, gradeable steps instead of one holistic verdict.
  • Writing precision. These are technical documents. Ambiguous phrasing produces unusable evaluation data.

Logistics

Fully remote, asynchronous, contractor engagement. Pay is per accepted task rather than hourly; the $40–90/hr band reflects rates observed across similar micro1 finance benchmarking projects and is not guaranteed — your effective rate depends on how efficiently you produce work that clears spec on first review. A weekly minimum submission count applies. micro1 fills these roles quickly and expects first tasks within 24–48 hours of onboarding, so treat this as an immediate-start commitment rather than a pipeline placement.