On Microsoft’s BizChat Analysis for Microsoft Teams, I label, validate, and correct datasets used to train AI models, and contributed to a 20% improvement in dataset accuracy through refined annotation workflows.
At Microsoft, I also labelled multi-turn conversational datasets for Semantic Machines’ conversational AI models. I helped define and document annotation standards, and tracked quality using Power BI and Excel/Power Query.
Before that, I delivered online English communication training at Talkeng Global & Clapingo Education and built evaluation rubrics for scoring student performance. I bring that experience applying consistent criteria to my data annotation work.
