AI Data & Content Specialist (Hybrid/Contract) at Hugo Technologies
I evaluated LLM-generated responses and search-based outputs for accuracy, relevance, and compliance with established guidelines. My work involved applying detailed rubrics for scoring safety, helpfulness, and adherence to instructions. I identified and documented edge cases, hallucinations, and reasoning gaps to support model improvements. • Maintained consistent QA scores of 95–98% in high-volume, fast-paced environments. • Collaborated with a global team to refine and enhance model behaviors. • Delivered structured feedback to inform model development and training. • Focused on complex multi-turn scenarios and documented nuanced behaviors.