Cantonese Prompt & Text Evaluators
Budget: €18 – €36 EUR
I’m assembling a small remote team of Cantonese-native linguists (living outside Mainland China) to help me evaluate AI-generated prompts and polish short, generic Cantonese texts. The work is 100 % online, task-based, and will be assigned in batches that you can complete at your own pace within agreed deadlines.
What you’ll actually do:
• Score and comment on Cantonese prompts produced by an LLM, flagging anything unclear, unnatural, or culturally off-base.
• Review short passages of everyday Cantonese, correcting spelling, grammar, register, tone, and factual slips. Your edits should both fix the issue and briefly explain why, so our engineers can feed the feedback back into the model.
I’m looking for linguists who have already dealt with translation/localization projects and who feel comfortable with AI-related tasks such as prompt engineering, content evaluation, or data labelling. Familiarity with annotation tools, spreadsheets, or CAT environments will help you start fast, but the specific platform training is provided.
Batches vary in size, so you can opt in or out depending on availability. All communication and file exchange happen via our secure portal; no VPN or Mainland China IPs can be used.
If you speak native-level Cantonese, enjoy fine-grained language work, and have experience giving structured feedback to AI or localization teams, I’d love to hear from you.
What you’ll actually do:
• Score and comment on Cantonese prompts produced by an LLM, flagging anything unclear, unnatural, or culturally off-base.
• Review short passages of everyday Cantonese, correcting spelling, grammar, register, tone, and factual slips. Your edits should both fix the issue and briefly explain why, so our engineers can feed the feedback back into the model.
I’m looking for linguists who have already dealt with translation/localization projects and who feel comfortable with AI-related tasks such as prompt engineering, content evaluation, or data labelling. Familiarity with annotation tools, spreadsheets, or CAT environments will help you start fast, but the specific platform training is provided.
Batches vary in size, so you can opt in or out depending on availability. All communication and file exchange happen via our secure portal; no VPN or Mainland China IPs can be used.
If you speak native-level Cantonese, enjoy fine-grained language work, and have experience giving structured feedback to AI or localization teams, I’d love to hear from you.