• Aug 2026 –Preferred Networks (Part-time Employee, following a summer internship)
• Apr 2026 –Meiji University, Graduate School of Advanced Mathematical Sciences, Master's Program in Frontier Media Science (expected completion: Mar 2028)
• Apr 2022 – Mar 2026Meiji University, School of Interdisciplinary Mathematical Sciences, Department of Frontier Media Science

Advances in computing and generative AI have made it possible for anyone to create a wide variety of things, such as software, images, and music, easily and quickly.

However, finding what is truly good for a particular person is not easy. The more things we can create, the larger the pool of candidates becomes, and no one can compare all of them alone. Moreover, when the image in your mind or your preferences are vague, it is difficult to clearly express what you want in numbers or words. Existing UIs, such as sliders that adjust one parameter at a time and chat interfaces where you give instructions in words, struggle to handle such vague requests, and the search stays within the limits of the user's own knowledge and imagination.

To address this, I research and develop systems that combine optimization techniques, which efficiently search a vast pool of candidates for what fits a person, including things they had not thought of, with interaction design that can handle vague images. My work spans many domains, including visual design, programming, 3D modeling, photo editing, and taste exploration. My goal is for people to go beyond the limits of their own knowledge and imagination and encounter truly good things that they could not have reached on their own. I call this "expanding a person's horizons," and I aim to achieve it through the systems I build.

In graduate school, I also focus not only on the result but on the process of searching. I believe that what people want is not fixed before they start searching, but takes shape as they search. Through searching, people encounter unexpected candidates and gradually broaden their perspective. I study ways of searching in which this process itself expands a person's horizons.

My motto is "Just build it." When I come up with an interesting idea, I give it shape first. This goes back to elementary school, when I started programming with Scratch and discovered the joy of creating worlds exactly as I imagine them.

I also value building things from scratch whenever I can. As the physicist Richard Feynman put it, "What I cannot create, I do not understand." I believe that I truly understand how something works only after building it myself. For example, I have built my own programming languages, including one for education and one for quantum computers. To understand how large language models (LLMs) work, I also trained a Japanese LLM from scratch on a single GPU in our lab. During my internship at Preferred Networks, I trained an image generation model from scratch using only rights-cleared images. Beyond optimization frameworks and interfaces, I can also build the machine learning models used inside them myself. I believe this lets me design the whole system so that users can reach more satisfying results with less effort.