Japanese Language Lead — Voice AI QA & Data Governance

Cartesia; United States; Remote; USD 34,000 - 83,000; Full time

Benefits offered by this job

Health Insurance

401(k)

Flexible PTO

Meals & Snacks

Job summary

Cartesia seeks native-level Japanese language resources to lead end-to-end language work for voice AI. You will own annotation quality, audit data, monitor product performance, and run projects requiring deep linguistic expertise.

This ongoing contractor role fits someone who can operate independently across workstreams. The engagement expects roughly 40 hours per week with flexibility for US Pacific hours to sync with the team and real-time support.

Qualifications

• Native-level proficiency in Japanese as spoken in Japan.
• Experience with data annotation, transcription, or language quality assurance.
• Strong foundation in linguistics or related field, with the ability to apply concepts like phonetics, prosody, and intonation when evaluating speech and writing annotation guidelines.
• Strong written English proficiency. All guidelines, feedback, reports, and team communication are in English.
• Ability to write clear, structured instructions and deliver specific, actionable feedback to annotators.
• Comfortable working with data and metrics such as inter-rater agreement scores, audit spreadsheets, and structured feedback templates.
• Self-directed and able to manage feedback cycles, QA reviews, and data projects without close oversight.

Responsibilities

• Answer questions from annotators about guidelines, edge cases, or language-specific issues via an agreed project communication channel.
• Cascade new feedback, instructions, or rubric updates to annotators working on your projects.
• Listen to annotator-rated audio files and check whether their ratings and written feedback align with the rubric.
• Prepare written quality assessments for annotators with specific examples and recommendations.
• Listen to TTS output and provide ground truth assessments of pronunciation, prosody, and naturalness in your language.
• Update or expand sections of the annotation guidelines when you find gaps or edge cases.
• Prep and run calibration meetings with annotators.
• Hold office hours on a mutually agreed schedule during active studies so annotators can ask questions in real time.
• Develop onboarding materials and resources for new annotators.
• Write QA reports on annotation quality and calibration outcomes.
• Review production audio in your language and flag quality issues to the team.
• Investigate language-specific issues reported by customers and identify root causes.
• Audit datasets for linguistic accuracy, completeness, and consistency.

Skills

Native Japanese

Linguistics knowledge

Data annotation experience

Transcription

Language QA

Strong written English

Instruction writing

Quality monitoring

Education

Linguistics or related field

Job description

Cartesia seeks native-level Japanese language resources to lead end-to-end language work for voice AI. You will own annotation quality, audit data, monitor product performance, and run projects requiring deep linguistic expertise.

This ongoing contractor role fits someone who can operate independently across workstreams. The engagement expects roughly 40 hours per week with flexibility for US Pacific hours to sync with the team and real-time support.