Evaluating LLMs for ADHD and Neurodivergent Workplace Support

Professor: Hossein Shirazi

Description: This project, conducted through the Digital Innovation Lab (DiLab), empirically investigates the capacity of large language models (LLMs) to generate context-sensitive, accessible, and psychologically attuned responses to scenarios involving ADHD and neurodivergence. The student will conduct systematic scenario-based evaluations, code response quality and failure modes, and examine model robustness under routine and high-stress conditions. The findings will contribute to an evidence-based framework for assessing inclusivity, reliability, and alignment in LLM-generated support.

Preferred Qualifications: Python programming, Interest in AI and deep learning, Analytical thinking, Effective communication

Contact Us