AI Model Evaluation & Structured Data Labeling Project
Worked on human-in-the-loop AI data labeling and evaluation tasks, including reviewing recording conversation with AI Model, conversational AI responses, validating function calls and parameters, identifying hallucinations, assigning severity levels, recognizing emotion, and ensuring guideline compliance. The work involved structured annotation of text and audio-based model outputs to improve model accuracy, reliability, and alignment.