AI Training Contributor
Contributed as a prompt and scenario author for LLM training projects on Outlier AI and CrowdGen. Generated, reviewed, and refined multi-turn tasks, long-form reasoning, and professional-context writing samples in English and French. Evaluated model outputs for factual accuracy, tone, and consistency, rewriting rubrics to improve reviewer alignment. • Authored scenario-based prompts and curated complex, real-world tasks targeting advanced model capabilities. • Conducted bilingual QA and performed rubric and instruction evaluations as part of the review cycle. • Worked in custom structured annotation platforms and read/edited JSON directly for specialized tasks. • Participated in invitation-only pilot programs due to consistently high performance and accurate evaluations.