Cantonese Language Evaluator

You will help improve AI systems that work in Cantonese by reviewing language data and evaluating model responses. The work covers both written and spoken Cantonese, with a focus on accuracy, fluency, context, and consistency.

• Assess Cantonese text and speech for accuracy, fluency, and contextual fit.
• Review AI model outputs and identify errors, unclear wording, and areas for improvement.
• Complete data annotation, labeling, and categorization tasks with digital tools.
• Apply evaluation guidelines consistently across different language samples.
• Share complex language cases with project leads and document useful findings.
• Provide recommendations that help improve evaluation processes.

This is an entry-level, part-time contractor role. Experience in language evaluation or related work is helpful but not required.

• Pay: $30-$40 per hour in USD.
• Time: 20 or more hours per week.
• Language: native or bilingual Cantonese, written and spoken.
• Location:
  
  available in the countries listed for the role; this is not a worldwide posting.
• Work style: independent work, multiple assignments, and project deliverables.
• Requirements: strong attention to detail, clear communication, consistent judgment, and comfort using digital review tools.
• Helpful background: language evaluation, translation, linguistic annotation, content moderation, transcription, or quality assessment.

AI training is the human work behind systems that understand language, images, audio, and other data. People review examples and model responses so AI can become more accurate, useful, and natural, and language specialists are paid for applying their knowledge carefully.