Moonlight MCQA - Chinese (Simplified) Generalist

$40–49/hrrate as listed

Write original general-knowledge multiple-choice questions in Simplified Chinese, each with ten answer options and a written solution, for use in an AI evaluation dataset. The work is remote and self-paced, with volume set by batch assignments rather than fixed shifts.

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

What the work involves

You draft exam-style multiple-choice questions in Simplified Chinese, pulled from areas you genuinely know — your graduate field, your professional domain, or general knowledge you can source and verify. Each item needs a clear stem, ten answer options (one correct, nine plausible distractors), and a worked solution that explains why the key is right and why the strongest distractors fail. The ten-option format is the hard part: it forces you to generate distractors that are wrong for distinct, non-overlapping reasons instead of padding the list with obvious throwaways.

Items are expected to be original — not translated from existing exams, not lifted from textbooks or question banks, and not reworded from material a model has likely already memorized. Reviewers check for ambiguity in the stem, more than one defensible answer, culturally or regionally unstable facts, and distractors that leak the key through phrasing or length. Expect revision requests on early batches while you calibrate to the style guide.

What the platform screens for

Credential verification — master's or PhD, plus 2–6 years of professional experience, checked against documents and your work history.
Language and context — native-level Simplified Chinese, and education or residence in China; the questions need to read as though written by someone inside that context, not translated into it.
Domain depth under follow-up — the screen will push past your first answer on a topic to see whether the expertise is real.
Item-writing judgment — whether you can spot what makes a question ambiguous, trivially guessable, or unanswerable without outside assumptions.

Logistics

Fully remote and asynchronous, with no set hours. Observed pay for this listing is $39.69–48.51/hr, typically tracked against submitted-and-accepted work rather than clocked time — bands are as observed on the platform and are not guaranteed. Contributors commonly work in blocks of a few hours, since drafting ten-option items with solutions is slower than it looks; a realistic first week is a small calibration batch, not high volume.

What they look for

Native-level Simplified Chinese (written)
Master's or PhD in any field
2–6 years professional experience
Educated or currently based in China
Exam or assessment item writing experience
Clear written explanation of reasoning

Before you apply

Would you pass this screen?

Drawn from what this listing actually requires. Answered in your browser — nothing is sent to us, and we do not store your answers.

Do you hold a master's degree or PhD in any field?
Master's or PhD · required
Since earning your degree, have you worked in a professional role for 2 to 6 years?
2–6 years professional experience · required
Were you educated (completed a degree) or are you currently based in mainland China?
Based in or educated in mainland China · required
Is Simplified Chinese your native language, or have you achieved native-level fluency in spoken and written Simplified Chinese?
Native-level Simplified Chinese · required

0 of 4 answered

Included with this listing

Interview prep — 17 questions

Matched to what this platform's screening actually measures. Answer each out loud in 60–90 seconds: claim → concrete example → implication.

What it tests: Whether the candidate understands pedagogical question design—specifically, can they engineer plausible wrong answers and avoid common pitfalls like making the correct answer too obvious or unintentionally signaling it through length or style.
A strong answer: A strong answer shows a deliberate process: identify the correct answer and its justification, then construct distractors that reflect real misconceptions or partial understandings in the domain, not random noise. The candidate should mention testing that no unintended cues leak through wording or option placement.
Where people lose it: Candidates who treat distractors as throwaway options or who admit they would simply list ten plausible-sounding statements without pedagogical intent will struggle here. The platform is buying rigorous question construction, not speed.