What the work involves
You read text and decide whether it is good, then explain why in writing that someone else can act on. In practice that means scoring model outputs for clarity, coherence and factual accuracy against a project rubric, editing documents for grammar, syntax, structure and tone, and flagging the specific markers of low-quality artificial text: the throat-clearing introductions, the triplet constructions, the confident filler that says nothing. Annotation tasks often ask you to label where a passage drifts from human register and to point at the exact clause, not the general vibe.
The written rationale is the deliverable as much as the score. Project coordinators use your comments to calibrate other reviewers and to feed training signal back into the model, so a one-line verdict is worth little. Some batches also ask for comparative judgments between two candidate texts, and some ask you to rewrite a weak passage so the contrast between poor and strong output is explicit.
What the platform screens for
micro1 runs an AI-led interview before any project match. Expect it to probe your actual editorial history (publications, document types, volume, who you reported to), your ability to articulate what separates competent prose from fluent emptiness, and whether you can hold a consistent standard across a long queue rather than drifting toward leniency by item forty. No AI background is required; the domain knowledge being tested is editorial. Preferred background includes a bachelor's in English, Linguistics or a related field, or prior experience teaching English, plus native-level English proficiency.
Logistics
- Remote, contractor engagement, fully asynchronous.
- Work is assigned in batches against defined deadlines; consistency and on-time delivery matter more than peak speed.
- Hours are typically flexible, and volume varies with customer demand rather than arriving as a steady weekly block.
- Pay band: $22–70/hr.