Korean–English Localization & Translation Expert (South Korea)

About OpenTrain

OpenTrain is a central job board for AI-training and data-labeling opportunities. We aggregate open positions from many AI companies and labeling platforms so contributors can find work in one place instead of hunting across dozens of sites.

Creating an OpenTrain account is free and applying takes only a few minutes. This listing is for a contract, part-time role supporting AI localization and linguistic quality assurance.

About AI Training Work

AI training (also called data labeling, annotation, or human feedback work) is the human side of building AI. For language roles that means reviewing, correcting, and evaluating machine-generated translations and localized text so models learn accurate, natural, and culturally appropriate language.

These projects are typically remote, flexible, and accessible to language professionals; they require familiarity with quality frameworks, tooling, and careful linguistic judgment.

The Role

We are hiring an experienced Korean–English localization and linguistic QA specialist to review and improve AI-generated outputs for the South Korea locale. This is a contract, part-time position with less than 20 hours per week.

You will evaluate accuracy, clarity, reasoning quality, and adherence to prompts and guidelines; make defensible LQA/MQM decisions; produce corrected text and model answers; and rate or rank multiple AI responses using structured criteria.

• Location: Must be currently based in South Korea (required).
• Employment: Contractor, part-time.
• Time: Less than 20 hours per week.
• Pay: PAY_PER_HOUR — USD $27/hour.
• Data type: TEXT; Label types: EVALUATION_RATING, RLHF, TRANSLATION_LOCALIZATION.
• Labeling software: OTHER; Experience level: Intermediate.

What You'll Do

Your day-to-day work will focus on evaluating AI outputs and producing high-quality localized text and explanations. Tasks are structured and evaluated against defined criteria and style guidance.

• Evaluate AI-generated Korean (and Korean↔English) outputs for accuracy, fluency, tone, and cultural appropriateness.
• Apply LQA/MQM error typologies to classify errors, assess severity, and make consistent decisions.
• Identify linguistic and methodology issues and perform focused fact-checking of units, names, dates, and references.
• Write corrected outputs, clear explanations, and model answers that show the preferred localization.
• Rate and rank multiple AI responses using structured evaluation criteria and provide defensible justifications.
• Use CAT tool workflows, terminology resources, TM/termbases, and QA checks as part of review.

Requirements

You must meet all of the following requirements to be considered for this project. We cannot accept applicants who do not meet these location, language, education, or experience standards.

• Must be currently based in South Korea (location requirement).
• Native or near-native Korean proficiency with strong professional writing/editing skills for the South Korea locale.
• English proficiency at C1+ (reading/writing) to interpret prompts, sources, and evaluation guidelines.
• Bachelor’s degree (or higher) in Translation, Linguistics, Localization, Communications, or a related field.
• Minimum 5 years professional localization/translation experience (multilingual localization preferred).
• Demonstrated experience applying MQM/LQA including severity/category decisions and consistent error classification.
• Proficiency with CAT tools and QA workflows, including TM/termbases and automated checks.
• Strong terminology management and copyediting ability covering tone, register, inclusivity, and cultural nuance.
• Ability to rigorously fact-check localized content (units, names, references, dates) using reliable sources and sound reasoning.

Preferred & Nice-to-Have

• Prior experience with AI data training, annotation, or evaluation is preferred.

How It Works / How To Apply

If you meet the requirements, apply through OpenTrain. Creating an account is free and the application takes only a few minutes.

Selected contributors will work as contractors, performing structured review tasks via the project's labeling/evaluation interface and CAT tools. Deliverables typically include rated responses, corrected text, and written explanations/model answers.

• Prepare a brief summary of your localization experience, MQM/LQA examples, and any CAT tools you use.
• Be ready to show examples or a short portfolio of localization/LQA work if requested during screening.