I am currently a Postdoctoral Scholar working with Dr. Emma Brunskill at Stanford University, affiliated with AI for Human Impact (AI4HI) Group and Stanford AI Lab (SAIL). I used to be a PhD student under supervision of Dr. Min Chi at North Carolina State University. I earned my M.S. under supervision of Dr. David M Whittinghill at Purdue University.
My research lies at the intersection of reliable AI, sequential decision-making, and human–AI collaboration. I am broadly interested in developing and evaluating AI systems, including LLM- and agent-based systems, that can interact with people safely and effectively, particularly in settings where decisions have long-term consequences.
A central theme of my work is decision-making under real-world uncertainty and empirical constraints, including limited or missing data, heterogeneous populations, distribution shifts, and imperfect feedback. My work spans human-centered domains such as healthcare and education, and increasingly connects these challenges to the evaluation, robustness, and safe deployment of AI-based systems.
My main research directions include:
- Develop and evaluate safe and robust AI-assisted decision-making for long-term human outcomes.
- Scale up ML to effectively serve diverse human populations and transferrable across contexts.
- Personalized interventions for enhancing individual outcomes.
I pursue these research goals in high-impact human-centered domains, particularly education (e.g., K-12, college, CS1/2) and healthcare (e.g., diagnosis and treatment for sepsis, Parkinson’s disease, HIV, etc.), where decisions directly shape long-term learning and health outcomes, and where reliable, personalized AI can meaningfully improve human well-being.
Contact me: ge.gao.hai AT gmail DOT com
- Ph.D., Computer Science, North Carolina State University
- M.S., Computer Graphics & Statistics, Purdue University
- B.E., Computer Science, Tianjin University
- Summer Session, University of California, Berkeley