About OpenTrain

OpenTrain is the #1 platform for starting and building careers in AI training and data labeling. We connect people with remote, flexible projects that teach AI how to understand language, images, audio, and model behavior. For this role, OpenTrain AI is the hiring and contracting organization.

About AI training and why it matters

AI training (also called data labeling or human feedback work) is the human side of building intelligent systems: people annotate, evaluate, and guide model outputs so models behave safely and usefully. These roles are remote, often flexible, and accessible — contributors directly shape how state-of-the-art models respond.

This job focuses on safety and policy evaluation for LLM outputs in both Korean and English, a high-impact area that helps reduce harm and improve model reliability across languages and cultures.

• 100% remote and contractor-based with flexible scheduling.
• Work directly influences model safety, policy alignment, and product risk decisions.

The role — what this job is

You will work as an AI Safety Data Reviewer / LLM Trainer, evaluating and labeling AI-generated text for safety, policy compliance, factual accuracy, and reasoning quality. This is an hourly, contractor, part-time role requiring 20+ hours per week and regular bilingual review across Korean and English.

Work will include evaluation ratings, question-answering checks, text-generation assessment, and RLHF-style review. Expect exposure to sensitive or disturbing content in a secure remote environment.

• Employment type: Contractor, Part-time.
• Hours: 20+ hours/week.
• Pay: USD $28–$38 per hour (typical rate $32/hr).
• Data type: Text. Labeling tasks: EVALUATION_RATING, QUESTION_ANSWERING, TEXT_GENERATION, RLHF.

What you'll do day-to-day

Your core responsibility is to judge model outputs against safety policies and provide clear, reproducible rationales that guide model improvements. You will evaluate multiple candidate outputs, spot conceptual or methodological errors, and recommend mitigations for risky behaviors.

• Review and label AI-generated Korean and English text for policy compliance and safety.
• Rate outputs on safety, factual accuracy, reasoning, and clarity.
• Identify edge cases, adversarial inputs, and failure modes during red-teaming.
• Supervise or advise on content-moderation decisions and policy application.
• Write concise, reproducible explanations that justify moderation or safety decisions.

Minimum requirements

Applicants must meet the following non-negotiable qualifications because this role demands bilingual policy judgement and senior-level safety experience.

• Near-native or native Korean proficiency in reading and writing.
• Minimum C1 English proficiency in reading and writing.
• Bachelor’s degree or higher in Communications, Linguistics, Psychology, Law/Policy, Security Studies, or equivalent professional experience.
• Senior-level experience in Trust & Safety, content moderation, policy operations, risk, compliance, investigations, or related safety functions.
• Proven LLM red-teaming or adversarial testing experience, including identifying edge cases and recommending mitigations.
• Strong knowledge of safety domains: hate/harassment, sexual content, self-harm, violence, bias, illegal goods/services, malicious activities, malicious code, and misinformation.
• Experience applying policy standards consistently across Korean and English content, including cultural nuance, slang, coded language, and context shifts.
• Strong analytical writing with clear, reproducible rationales.
• Comfortable handling explicit, toxic, violent, sexual, or psychologically disturbing content in a secure remote work environment.

Preferred qualifications

These are not required but will make you more competitive and effective in bilingual safety review and localization-sensitive evaluation.

• Localization or translation experience with ability to preserve meaning, severity, and intent across languages.
• Prior experience with RLHF workflows, evaluation rating systems, or annotator training.
• Experience writing or contributing to moderation or safety policy documentation.

How the work is structured

Tasks are assigned through OpenTrain's workflow and are completed remotely in a secure environment. You will receive task guidelines, policy documentation, and examples; your outputs will be reviewed and may be used to refine policies or model behavior. Compensation is hourly and paid on a contractor basis.

• You will follow detailed instructions and example-driven rubrics for each task.
• Expect both individual review tasks and collaborative red-team sessions or calibration checks.
• Secure handling of sensitive content and adherence to confidentiality requirements will be required.

How to apply and next steps

Create a free OpenTrain account, complete your profile with language skills and relevant experience, and submit an application for this AI Safety LLM Trainer role. Shortlisted candidates will be invited to a skills check and policy calibration exercise to confirm bilingual judgement and red-team capabilities.

• Include details of your Trust & Safety or moderation experience and examples of red-teaming work if available.
• Be prepared for a timed or practical evaluation in Korean and English focused on safety reasoning and writing.