The work

You draft original multiple-choice items from your own practice knowledge — not from textbooks, exam banks, or scraped past papers. Each item presents a legal problem grounded in Taiwanese or Hong Kong law, offers ten answer options, and is accompanied by a worked solution that explains why the correct answer holds and why the strongest distractors fail. Everything is written in Traditional Chinese, with statutory citations, case names, and terminology in the form a local practitioner would actually use.

The difficulty target matters more than volume. Items that a competent model answers correctly on first read contribute little; the goal is questions that require multi-step reasoning, correct identification of the governing provision, or awareness of a distinction that a generalist would miss — for example, the divergence between Hong Kong common law positions and PRC-influenced statutory instruments, or Taiwanese civil code interpretations that differ from Japanese or German antecedents. Distractors should be plausible to a trained lawyer, not obviously wrong.

What the screening looks for

  • Verifiable credentials: a law degree or the national equivalent, and education or practice grounded in Taiwan or Hong Kong. Expect to name your institution, qualification route, and practice areas.
  • Genuine domain depth: follow-up questions probe specific provisions, procedural rules, and how you would resolve an ambiguous fact pattern. Surface familiarity is easy to detect under a second or third question.
  • Language precision: native-level Traditional Chinese legal writing, including correct register and jurisdiction-appropriate terminology.
  • Evaluation judgment: whether you can tell a hard question from a merely long one, and whether your solutions defend a single defensible answer rather than gesturing at complexity.

Logistics

Fully remote and asynchronous. Hours are flexible with no fixed schedule; most contributors work in blocks and submit batches against a task specification with quality review. Observed pay for this listing is $39.69–48.51/hr, set by the platform and not guaranteed — rates vary with assessed expertise, task type, and project stage. Engagements are typically contract-based and scoped by project, so duration depends on dataset needs.