Litao Yan
Litao Yan

Hi, I'm Litao.

I am a Ph.D. candidate in Computer and Information Science at the University of Pennsylvania, advised by Andrew Head. I expect to graduate in May 2027. I am currently on the job market for research and engineering roles starting in 2027.

My research lies at the intersection of AI agents, developer tools, and AI evaluation. I build AI systems with new agent architectures and interaction techniques for understanding and inspecting AI-generated work, particularly in AI-assisted programming. I also develop task datasets and benchmarks to evaluate these systems through technical validation and studies of human understanding and use.

I have also worked on multi-agent systems and agent evaluation at Adobe Research, and multimodal AI systems at Microsoft Research. Previously, I received my M.E. in Computational Science and Engineering from Harvard University, where I worked with Elena Glassman on ExampleNet, which received a CHI Honorable Mention Award.

News

Submitted our latest work on execution-grounded review of AI-generated code to CHI 2027.
Finished my research internship with the Evaluation and Experience Lab at Adobe.
InvisibleMentor was accepted to CHI 2026.
Trailblazer was presented at UIST 2025.

Publications

Revealing Program Behavior with Execution Digests

Litao Yan, Jeffrey Tao, Daniel Fried, Andrew Head

Under review