Haruki Yajima

I am Haruki Yajima, a first-year master’s student at The University of Tokyo. I am enrolled in the Department of Information and Communication Engineering at the Graduate School of Information Science and Technology.

My research interests lie in AI safety and trustworthy AI. I currently work as a research assistant with Prof. Isao Echizen on AI safety. As an undergraduate, I worked with Prof. Yusuke Matsui on reliable pruning methods for machine learning models. This work was accepted to ICML 2026 (CORE: A*).

I am also interested in bringing AI technologies into real-world applications. At Supercell AI Innovation Lab, with support from Supercell, I explore new game experiences powered by AI. As an ML engineer intern at MinD in a Device, I worked on inspection and measurement applications using 3D reconstruction technologies.