On independent projects and platform work, I compare text and image AI outputs against project guidelines, checking accuracy, helpfulness, safety, and prompt adherence.
I evaluate AI-generated images for quality, realism, visual artifacts, and anatomical consistency, and review text-and-image prompts to check that outputs match the requested details.
I also research and verify real-world information to catch potential hallucinations, then document visual errors and refine prompts to assist with model improvement.
