Search Relevance & AI Response Auditor (UHRS)
Focus on evaluating Large Language Model (LLM) responses and search engine snippets for relevance and factual accuracy. Responsibilities include analyzing bilingual (Chinese/English) search intents, verifying information against reliable sources, and labeling datasets to improve the ranking algorithms of major search engines. Adhered to strict quality benchmarks (Spam Accuracy > 85%) under high-volume production environments.