Behavioural scientist studying how AI systems behave toward the humans who use them. I design and run empirical evaluations of AI conversational behaviour, including over-delivery, intention override, and autonomy erosion, and translate findings into concrete behaviour specifications and design interventions. PhD researcher at UCL Interaction Centre (UCLIC).
I work at the intersection of behavioural psychology, AI design, and product strategy.
My PhD at UCL investigates how conversational AI systems, optimised to be maximally helpful, end up overriding the intentions of the people using them, and what that costs teenagers still developing the judgment to know the difference.
That question has made me a sharper researcher, a more rigorous designer, and someone who understands why 'helpful' products can still fail the people they're meant to serve.
Beyond research: Kathak practitioner for 14 years. Published poet. 2nd Prize at IIT Mandi's international research conference.
"When an AI system is optimised to be maximally helpful, it can quietly do the thinking, deciding, or feeling for you instead of with you. For a teenager still learning to trust their own judgment, that's not a convenience. It's a design problem. And design problems have design solutions."
EMA
Diary Studies
Usability Testing
Figma
Qualtrics
User Interviews
Product Strategy
R
Prototyping
LLM Evaluation
AI Ethics
Human-AI Interaction
