Chinese Data Annotation Project

I have a dataset in Chinese that now needs meticulous human annotation so it can be fed into downstream AI experiments. The exact modality (text, image, or short audio) and tag set will be shared once we start; for now, think of typical tasks such as marking sentence boundaries, labeling named-entities, classifying sentiment, or assigning scene tags on photos.

What matters most is that every label you apply is both linguistically correct and consistent with the guidelines I will supply. Native-level Mandarin (Simplified) reading ability is essential, and familiarity with any mainstream annotation platform—Label Studio, Prodigy, Excel/CSV sheets, or similar—will make the workflow smoother.

Deliverables
• Fully annotated files in the agreed format (JSON, CSV, or platform export)
• A concise changelog explaining any edge-case decisions
• Daily progress updates so I can run random spot-checks

Acceptance criteria
• ≥ 95 % agreement with my gold-standard sample on the first batch
• Clean file structure: no broken encoding, no trailing spaces, all IDs preserved

Once we align on scope and tool preference, I’ll open the first tranche of the data and timeline so you can get started right away.