My research is about building AI systems that help people discover, reason, write, learn, and reflect.
I am a PhD student in Computer Science at the University of Toronto. My current interests include AI for science, computational creativity, language models, knowledge representation, and human-computer interaction. A recurring theme in my work is how AI systems can support useful human thinking, rather than only automate the final answer.
Research perspectives
AI for science and computational creativity
I am interested in AI systems that can support scientific discovery: generating hypotheses, planning experiments, organizing knowledge, and helping researchers explore ideas that would be difficult to search manually. To me, computational creativity in science is not just about producing surprising outputs, but about making the scientific process more adaptive and reflective.
Language models, schemas, and reasoning
Another direction of my work studies how language models use structure. I am interested in schemas, in-context learning, and representations that make AI reasoning easier to inspect, reuse, and improve. This connects to my broader interest in building AI systems that are not only powerful, but also more legible to the people who use them.
Human-AI tools for reading and writing
I have worked on systems that help people read and write long-form academic documents with AI support, including hierarchical paper reading and planning tools. This line of work asks how AI can help people navigate complex information without flattening the user’s own thinking.
Learning systems and adaptive experiments
Earlier, I worked on adaptive experimental infrastructure for education, including work connected to the XPRIZE Digital Learning Challenge. I am interested in systems that allow learning environments to improve continuously while still respecting the practical constraints of instructors, students, and researchers.
Reflection, voice, and HCI
My HCI work has also explored reflection tools, voice-based interaction, and lightweight interventions in learning or well-being contexts. I care about interfaces that lower friction and help people express what they are thinking, especially when typed interaction is not the best medium.
For a more complete and current list of publications, please see my Google Scholar profile.