About Me
I am a Principal Researcher at Microsoft on the Excel Team, where I work on human-AI collaboration in Agent Mode. My research focuses on how people collaborate with AI, examining how individual differences, task contexts, and system design shape trust, reliance, and decision-making. I've investigated user mental models of AI agents, repair approaches in conversational breakdowns, and the impact of AI identity disclosure and communication directionality on collaboration outcomes.
I've worked across conversational AI, data annotation systems, and LLM evaluation, most recently leading the research agenda behind EvalAssist, a human-centered LLM-as-a-Judge platform that helps practitioners build, refine, and scale evaluation criteria through interactive workflows.
Before Microsoft, I spent nearly a decade at IBM Research driving the human-AI collaboration agenda. I hold a Ph.D. in Human-Computer Interaction from the University of Maryland, College Park, where I also completed internships at IBM Research (Almaden), Microsoft Research (Cambridge, UK), and Data Science for Social Good.
