Kasia Kolesinski

I combine a background in special education with hands-on prompt engineering, bringing human-centered insight to AI development. Teaching children with diverse learning needs taught me to break down complex ideas, personalize language and communicate with clarity.

In my current work I design and refine prompts across text, image and video, review and annotate high volumes of model outputs, and label edge cases to flag hallucinations and safety issues. I have created documentation, synthesized findings into actionable reports, and contributed to measurable reductions in error rates while improving model behavior.

I thrive in collaborative, experimental environments and apply rigorous data analysis, linguistic precision and adaptive troubleshooting to improve model performance and user experience. I am motivated to translate educational assessment and individualized planning skills into reliable, ethical AI systems.