Japanese PDF annotation and transcription expert mapping document structure and faithfully transcribing published pages. Creating human - authored training data for Mercor’s frontier AI partners.
About the role
- Japanese PDF annotation expert mapping layouts and transcribing documents for Mercor, which partners with AI labs to train frontier models. Reviewing Japanese-language training data for accuracy.
Responsibilities
- Find publicly accessible Japanese PDFs in assigned document types and record their sources
- Identify and bound every meaningful page region
- Assign component types and reading-order indices
- Link regions to their associated figures or tables using parent component identifiers
- Transcribe all text exactly as shown, including Japanese scripts, furigana, and handwritten content
- Flag illegible source regions
- Capture language, document type, source, page dimensions, and table/formula/handwriting metadata
- Review colleagues' work end to end as a second Japanese expert
- Produce human-authored training data for document AI
- Ensure layout diversity and accurate representation of Japanese documents
Requirements
- Native Japanese speaker with full command of kanji, hiragana and katakana, including furigana and variant character forms
- Professional experience in interpretation, journalism, transcription, translation, editorial work, or comparable document-intensive work
- Character-level accuracy and attention to detail
- Ability to apply a taxonomy consistently across hundreds of pages
- Comfortable with vertical text, multi-column newspapers, exam papers, and handwritten forms
- Fluent language skills required: Japanese
- Ability to find publicly accessible Japanese PDFs in assigned document types containing at least one multimodal element
- Ability to identify, bound, type, and order document regions
- Ability to transcribe kanji, hiragana, katakana, furigana, and handwritten content faithfully
- Independent contractor status
- Must not require H1-B or STEM OPT sponsorship
- Nice-to-have: annotation, labeling, grading, or bilingual evaluation for AI training data
- Nice-to-have: MTPE, subtitling, bilingual QA, OCR correction, or post-editing
- Nice-to-have: typesetting, copy-editing, proofreading, or digitization of Japanese-language material
- Nice-to-have: Unicode normalization, Japanese input methods, full-width and half-width forms, and kanji variant handling
Benefits
- Fully remote role
- Flexible own schedule
- Weekly payments via Stripe or Wise
- Projects may be extended, shortened, or concluded early depending on needs and performance
- Competitive payment
- Collaborate with leading researchers
- Help shape the next generation of AI systems
- Referral program offering up to $140 per successful referral
Job title
Job type
Experience level
Salary
Degree requirement
No Education Requirement
Location requirements
Report this job
Found something wrong with the page? Please let us know by submitting a report below.
Japanese PDF annotation expert mapping layouts and transcribing real documents. Creating high - accuracy training data for Mercor’s document AI and frontier - model projects.
Japanese PDF annotator mapping page structure and transcribing Japanese documents. Creating human - authored training data for Mercor’s document AI and frontier model projects.
Japanese PDF annotation expert mapping layouts and transcribing real documents for Mercor’s AI training projects. Ensuring character - level accuracy across complex Japanese pages.
Freelance Japanese Teacher delivering personalized online lessons for Busuu’s global language - learning platform. Teaching corporate clients through 1:1 coaching, needs analysis, and bespoke learning paths.
Japanese PDF annotation and transcription expert mapping document structure and transcribing complex pages. Creating human - authored training data for Mercor’s frontier AI systems.
Japanese PDF annotation and transcription expert mapping page structure and transcribing text for Mercor’s AI training datasets. Fully remote independent contract work focused on high - accuracy Japanese documents.
Japanese PDF annotation and transcription expert creating human - authored training data for Mercor’s document AI projects. Mapping document structure and transcribing Japanese text character by character.
Japanese PDF annotation and transcription expert mapping document structure and faithfully transcribing text. Creating human - authored training data for Mercor’s frontier AI projects.
Japanese PDF annotation and transcription expert creating training data for Mercor’s document AI projects. Structuring pages and transcribing Japanese text for frontier AI models.