AI Safety Evaluator - Korean - Remote

Overview

Welo Data is seeking experienced bilingual evaluators for a multilingual AI safety project that focuses on evaluating model responses across culturally specific prompt-image datasets.

In Short

• Evaluate AI-generated responses using a structured safety rubric.
• Complete two independent evaluations per item.
• Document evaluations primarily in English.
• Participate in calibration sessions.
• Support arbitration for evaluation discrepancies.
• Maintain quality and throughput targets.
• Fluency in the target language and English required.
• Strong analytical skills and attention to detail.
• Freelance role with flexible hours.
• Comfortable working with explicit content.

Requirements

• Fluency in the target language and English.
• Deep cultural understanding of the target locale.
• Strong written English skills for documentation.
• Prior experience in safety evaluation preferred.
• Ability to apply guidelines consistently.
• Reliable availability during the production window.
• Experience with prompt-image evaluation projects is a plus.

Benefits

• Work remotely from South Korea.
• Flexible schedule options available.
• Competitive hourly rate of 22 USD.
• Opportunity to work with a global team.
• Gain experience in AI evaluation and safety assessment.