AI Data Annotator & LLM Content Evaluator at Appen — Crowdgen Platform

Evaluated AI model outputs and annotated training data across text categorization, sentiment analysis, and multi-turn conversational tasks; designed real-world prompts; assessed model responses for grounding, contextual relevance, helpfulness, and hallucination mitigation; produced structured reasoning traces and written quality rankings across multiple model versions with analytical justification.

UHRS Judge / LLM Response Evaluator at Microsoft (UHRS)

Assessed AI-generated content, search results, and conversational model responses for quality, intent alignment, and policy compliance using UHRS evaluation frameworks; evaluated multi-turn query flows and entity-tagging tasks; provided detailed written assessments and feedback to inform LLM performance improvements.