AI Data Annotation Contractor | Outlier (Remote)
I evaluated and annotated AI model outputs for Outlier by applying comprehensive rubrics. This work focused on accuracy, helpfulness, and safety of text generations. My analysis supported improvements in AI reasoning and instruction adherence. • Produced structured feedback on diverse prompt categories. • Reviewed model-generated responses for factual correctness and utility. • Assessed adherence to safety and user instructions in generated content. • Supported the iterative improvement of LLM capabilities through detailed assessments.