Bilingual AI Data Annotator – Traditional Chinese

TaskUs; Singapore; On-site; SGD 29,000 - 42,000; Full time

Job summary

TaskUs is seeking a meticulous AI Data Annotator in Singapore to annotate voice and text conversations for a multilingual project. You will contribute to the development of next-generation AI by accurately labeling data to improve model performance and safety.

The role requires fluency in Traditional Chinese and English, strong attention to detail, and the ability to work with annotation tools. This position emphasizes quality, independence, and timely delivery on project deadlines.

Qualifications

• Fluent in Traditional Chinese and English with strong grammar and comprehension.
• High attention to detail with accuracy over repetitive tasks.
• Comfortable using digital data annotation platforms and tools.
• Excellent time management and organizational skills.
• Prior experience in data labeling, annotation, or transcription is a plus.

Responsibilities

• Text Annotation: label text conversations per guidelines, identify entities, sentiment, and intent.
• Voice Annotation: transcribe speech, identify speaker changes, and label conversation parts.
• Apply clear, consistent labels to text and voice data to ensure quality.
• Follow project-specific instructions to meet quality standards.
• Work independently and manage time to meet deadlines.

Skills

Chinese-English fluency

Attention to detail

Time management

Organizational skills

Data labeling/annotation experience

Tools

Data annotation tools

Job description

TaskUs is seeking a meticulous AI Data Annotator in Singapore to annotate voice and text conversations for a multilingual project. You will contribute to the development of next-generation AI by accurately labeling data to improve model performance and safety.

The role requires fluency in Traditional Chinese and English, strong attention to detail, and the ability to work with annotation tools. This position emphasizes quality, independence, and timely delivery on project deadlines.