Founder & CEO (AI Data Labeling, Content Review Operations)
Designed and operated a narrative risk and content review system for labeling and categorizing online text data. Applied a detailed harm taxonomy and policy guidelines to annotate user-generated content for social harms. Reviewed and documented edge cases to support moderation system training and evaluation. • Used internal/proprietary tooling and productivity platforms as labeling and documentation infrastructure. • Regularly leveraged large language models for prompt engineering, draft evaluations, and content analysis. • Created written procedures so reviewers could consistently perform detailed text content labeling and escalate high-risk items. • Balanced speed and accuracy while handling continuous, high-ambiguity content labeling queues.