Mandarin Quality Control Lead

About the role

You will own the end-to-end quality strategy for the Mandarin Chinese locale within a large-scale AI training program, defining and driving standards that ensure linguistic precision and task accuracy. In this capacity, you will lead a team of Quality Control Reviewers, orchestrating their workflow and aligning their evaluations with strict project guidelines to maintain consistency. You will serve as the central authority for quality interpretation, resolving disputes and clarifying requirements to uphold the integrity of the data pipeline. The role requires you to design and deliver structured coaching sessions that elevate reviewer performance and reinforce guideline comprehension. You will analyze performance data to detect emerging risks, implementing corrective actions that prevent defects from propagating downstream. Collaboration with global peers will position you as a driver of process improvements that scale across language teams and AI initiatives. You will also support the systematic onboarding of new reviewers, ensuring they are fully equipped to execute tasks at the required quality level. Ultimately, your work will directly influence the reliability of AI systems by safeguarding the linguistic quality of Mandarin Chinese training data.

Key facts

What you'll do

Establish and refine quality standards for Mandarin Chinese tasks, translating high-level guidelines into actionable criteria for reviewers.

Conduct detailed audits of reviewer outputs, identifying patterns of deviation and enforcing alignment with project specifications.

Lead calibration sessions that dissect complex examples, fostering shared understanding and consistent application of quality rules.

Provide evidence-based feedback to individual reviewers, highlighting strengths and specifying concrete improvements.

Monitor key quality metrics, generating insights that inform training priorities and process adjustments.

Escalate intricate quality issues to relevant stakeholders, proposing viable resolutions that balance accuracy and efficiency.

Develop and maintain documentation that clarifies guidelines, examples, and quality expectations for the Mandarin locale.

Collaborate with engineering and product teams to incorporate feedback into workflow optimizations that reduce ambiguity.

Support the evaluation of new tasks, assessing clarity, cultural relevance, and feasibility for the target language.

Mentor junior reviewers, facilitating their growth through structured coaching and performance tracking.

Partner with global Quality Control Leads to harmonize practices across languages and ensure cross-regional consistency.

Analyze edge cases, contributing to the evolution of guidelines and reference materials for ambiguous scenarios.

Track reviewer workload and quality trends, proactively addressing bottlenecks that could impact project timelines.

Contribute to the selection and onboarding of new Quality Control Reviewers, defining assessment criteria and training materials.

Drive continuous improvement initiatives that enhance documentation, tooling, and communication within the quality function.

Requirements

Demonstrate native-level proficiency in Mandarin Chinese, as defined by CEFR C2 or an equivalent standard.

Exhibit professional written and spoken English, enabling clear communication with global stakeholders.

Bring a minimum of two years of experience in AI training, data annotation, linguistic quality, or a related quality assurance discipline.

Possess at least one year of practical experience leading or supervising reviewers, annotators, or a quality-focused team.

Showcase a strong grasp of linguistic quality concepts, including grammar, fluency, tone, register, and cultural nuance.

Have a proven ability to deliver structured, evidence-based feedback that helps others improve their performance.

Exhibit strong analytical skills, particularly when evaluating tasks where multiple interpretations are valid and a single correct answer is absent.

Display robust organizational skills, managing multiple priorities and deadlines in a fast-paced remote work environment.

Maintain attention to detail, ensuring that quality checks are thorough and consistently applied across all assignments.

Demonstrate ownership of processes, taking initiative to identify gaps and propose enhancements without direct supervision.

Nice to have

Hold a degree in Linguistics, Computer Science, Engineering, or another analytical discipline.

Have hands-on experience with Large Language Models, RLHF, AI data annotation, prompt evaluation, or NLP projects.

Familiarity with annotation platforms and quality management processes specific to AI datasets.

Practical notes

Commitment: 30-40 hours per week.

Job Type: Fixed-term contract.

Location: Japan, Remote.

Start Date: ASAP.