At Meridial Marketplace, I train and evaluate generative AI models through audio evaluation, transcription review, and quality assurance. I rate responses in English and Spanish against rubrics and check outputs against detailed guidelines.
At Outlier AI (Scale AI) and Mindrift (Toloka), I evaluated audio, visual, and text outputs, including side-by-side comparisons of LLMs. Earlier, as a Jr. RPA Developer at Banco Ficohsa, I developed UiPath workflows and trained and tested Python OCR models for banking documents.
