AI Training & Evaluation – Project Diamond (Handshake AI)
In Project Diamond at Handshake AI, I was responsible for annotating and evaluating AI-generated responses. My work ensured model outputs were consistent and aligned with established guidelines, directly impacting overall model quality. I contributed to refining the annotation workflow and advancing feedback mechanisms for improved performance. • Conducted detailed analysis of AI-generated text outputs • Annotated and scored responses based on strict criteria • Ensured data quality and compliance with project guidelines • Assisted in developing better annotation protocols