At Invisible Technologies, I train and evaluate large language models through image-based evaluations, multimodal tasks, red-teaming, and complex research campaigns.
I provide actionable feedback to improve model quality, factual accuracy, safety, and performance. My work includes SFT, RLHF, prompt-response engineering, and text-to-image alignment.
My psychology background informs a human-centered approach to model evaluation, including prompt-fidelity scoring, visual artifact isolation, image captioning, and synthetic data curation.
Alongside AI training, I’ve worked in freelance styling and photography for editorial and commercial clients, including Ocean Drive Magazine and Avessa Magazine. I also hold a Bachelor of Arts in Fashion Styling from Istituto Marangoni.
