AI Quality Analyst (Gemini): Chinese
- Pay
- $15 / Hour
- Open to
- Worldwide
- Apply
We earn a commission if you sign up through the links on this page. It costs you nothing and does not affect which jobs we list. How this works.
- Skills
- ai evaluation
- chinese
- side-by-side ranking
- prompt design
- annotation
- written rationales
What you'll do
Chinese is the focus language of this Gemini personalization evaluation. You judge whether the model uses what it knows about you (past Gemini conversations, Gmail, Search and YouTube activity) in ways that are accurate, natural and useful.
The loop is simple:
- Design multi-turn starting prompts (typically 1 to 5 turns) based on your own context
- Flag grounding problems: claims about you without evidence, or flawed inferences
- Compare two responses side by side and rank them for helpfulness, ease of use and enjoyment
- Write rationales that cite exactly which turns showed an issue or a strength
- Verify "Debug Info" to confirm the chat summaries and data sources were used
- Delete evaluation conversations afterwards to keep your chat history clean
Compared with the Portuguese version of this project, this ad is shorter and lists fewer logistics: it does not mention using your primary personal Google account or full-time local-time availability, though the task description clearly relies on your own data.
Who fits
- High-level reading and writing in Chinese (the ad does not say which variety or script)
- Creativity in prompt design and a feel for forced connections and poor inferences
- Close attention to naturalness differences between near-identical answers
- Clear, structured written rationales
- A BS/BA or equivalent experience in a relevant field (policy, law, ethics, linguistics, journalism, computer science); data annotation, AI evaluation or content moderation experience is strongly preferred
What it pays
The ad states $15 per hour.
Worth knowing
Good:
- Published hourly pay
- No stated residence requirement
- Focused, well-defined work
Less good:
- Evaluation uses your own account data, a real privacy trade-off
- At least 4 hours a day and up to 40 hours a week, with 4 hours of overlap with PST
- Three-month contractor engagement
- Profile review, then a 24-hour assessment, then pre-onboarding discussions for shortlisted candidates
About this listing
Posted on Turing's board on 16 February 2026 and read on 18 October 2026. Rate and engagement terms are Turing's own text. See Turing.
More roles at Turing
See all 118Similar roles at other platforms
Browse similar roles
Not the right fit?
See every open role, or get new ones on Telegram or Discord as they are added.