I’ve worked on labeling datasets for chat models mostly tagging conversations for intent, tone, and safety issues, plus
I’ve worked on labeling datasets for chat models mostly tagging conversations for intent, tone, and safety issues, plus rating model responses for accuracy and helpfulness. I’m comfortable following annotation guidelines, spotting inconsistencies, and doing quality checks on my own work and other labelers’ work. I’ve also given feedback on where the model was failing so the team could adjust the prompts and guidelines. It’s not expert-level, but I can handle most standard labeling tasks and understand how the labeled data feeds back into improving the model.
Not specified