What the work actually involves

You will spend most of your time reading candidate text — model outputs, drafts, or paired responses — and making a judgment call with a written rationale attached. Some tasks ask you to score a passage against a rubric for clarity, coherence, factual accuracy, and tone. Others ask you to annotate specific spans: the hedging preamble, the tricolon that says nothing, the false-balance conclusion, the confidently wrong citation. A third type asks you to edit rather than score, producing a corrected version that a model can learn from, and to explain in a sentence or two what the original got wrong. Throughput expectations are real: rubrics evolve mid-project, and consistency across a batch matters as much as brilliance on any single item.

What the screen is looking for

micro1 runs an AI-led interview before any project assignment. It is not testing whether you have worked in AI — the listing is explicit that domain knowledge is what counts. It is testing whether you can articulate editorial judgment out loud. Expect follow-ups that push on a claim you just made: if you say a passage is incoherent, you will be asked which sentence and why. Vague praise for "good writing" or an inability to distinguish a style preference from an actual error is where most candidates lose ground.

  • Concrete editing history — publications, teaching, QA work, style guides you have worked under
  • The ability to name specific markers of machine-generated text rather than gesture at "it just sounds off"
  • Rubric discipline: scoring what the instructions ask for, not what you would personally have written
  • Comfort flagging ambiguity upward instead of silently guessing

Logistics

Fully remote, contractor engagement, largely asynchronous. Assignments arrive with deadlines and you work through them on your own schedule, with occasional syncs with project coordinators. Australia / Asia-Pacific availability is a stated preference for coverage reasons, not an absolute bar. Volume fluctuates with customer demand — treat this as a supplementary engagement rather than a guaranteed weekly load, and note that rates vary by project and assessed skill level within the observed band.