At Outlier, I evaluated and annotated AI-generated outputs for foundational language models, maintaining 94–96% quality scores through spot audits and lead-reviewer evaluations. I worked across prompt engineering, RLHF text ranking, multi-turn reasoning, and factuality verification, documenting edge cases to support evaluation improvements.
At Scale AI, I applied taxonomies to text and image datasets and achieved 95% inter-rater consistency and audit pass rate across assigned queues. My freelance design and video projects included reviewing interfaces for usability and editing footage for timing, sequencing, and continuity.
