At DataAnnotation.Tech, I evaluate and edit Korean chatbot responses for accuracy, instruction following, localization, and fluency. I create evaluation criteria and prompts, translate English prompts and responses into Korean, and test model responses to sensitive and adversarial prompts against safety guidelines.
My translation and interpretation work has included translating a policy research report for ASEM Global Ageing Center and interpreting for the Hyundai Capital NGS Project across IT and finance streams. I’ve also evaluated AI translations and English speech recordings in the NIA AI Training Data Development Project.
