Adnan Abbas

I am a human–computer interaction (HCI) researcher and fourth-year PhD student in Computer Science at Virginia Tech, where I am advised by Dr. Sang Won Lee in Echolab.

I design and study proactive, agentic AI systems that support planning, self-reflection, and behavior change. I am particularly interested in conversation as a medium for reflection and in how our interactions with intelligent systems can shape our behavior and everyday practices.

My research focuses on:

• Designing agentic systems that act on external environments, maintain context across multi-turn conversations, and adapt to user preferences.
• Understanding how people negotiate and calibrate the autonomy and initiative of goal-driven agents for behavior change.
• Studying long-term human–AI adaptation and its implications for collaboration, trust, and knowledge work practices.

I use human-centered methods to design, build, and evaluate interactive AI systems. I conduct user studies through controlled experiments and field deployments, using qualitative and quantitative analysis to understand behavioral outcomes and perceived value, then use these findings to refine system design.

During my research internship at Toyota Research Institute, I worked with behavioral scientists to develop and evaluate an AI tool for diagnosing behavioral barriers and designing interventions at scale.

Before my PhD, I worked as an Educational Technology Engineer at Educative, developing interactive courses on WebXR and A-Frame and VR development with Unity. As an undergraduate at LUMS, I worked with Dr. Zafar Ayyub Qazi on optimizing the design of control planes in cellular networks.