Start with the kind of opportunity
A job title is only the beginning. A published role may describe a current project, an ongoing intake or a route into an expert network. A talent pool collects interested candidates for possible future assignments. Treat these as different decisions: applying to a pool is not the same as accepting a funded project.
Read the actual assignment
Look for what you will produce: code reviews, model evaluations, language ratings or domain feedback. Then check the expected format, deadlines, equipment and whether the work includes qualification tests. A broad title such as “AI expert” tells you less than a clear description of the tasks. Our short summaries help you navigate; the original platform remains the authority.
Separate a rate from earnings
A stated hourly rate is not a promise of available hours. A per-task price cannot be compared with an hourly rate until you know how the task is defined and paid. Keep currency and unit together. Ask whether assessment time, revisions and rejected submissions are compensated when the source does not say. Do not multiply a headline rate into an annual salary.
Check location before investing time
Remote describes where work can happen, not which countries a platform supports. Read residency, language, identity verification and payout requirements. Even a worldwide network may restrict individual projects. If location evidence is missing, treat eligibility as unverified instead of assuming you qualify.
Use freshness as evidence
A recent check means the record was visible when checked. It does not tell you when the employer posted it or whether selection is still active. Paused or failed sources need extra confirmation. Open the official link before spending time on an application, and keep a copy of the project terms you actually agree to.
Sources and further reading
Platform requirements change. Always confirm them on the official listing.