Hello! I’m a master’s student in Data Science & Mathematics at EPFL, concentrating on multi-agent systems.
Currently, I’m collaborating with the AI Research Center at AIST. Together, we conduct research on automating research and business ideation using LLM agents, and on social simulations that address social-scientific questions.
My work spans social simulation with LLM agents and medical machine learning. My goal is to build computational tools that foster constructive human dialogue and help bridge divisions in democratic societies.
Research Interests
Education
Research Experience
SPRING Lab, EPFL
Research Project (Supervisors: Prof. Carmela Troncoso, Dr. Ana-Maria Cretu)
Aug 2024 – Jan 2025 · Lausanne, Switzerland
Explored perceptual-hash inversion with diffusion models to study privacy impacts in computer vision systems.
AI Research Center, National Institute of AIST
Technical Staff at KIRT (Supervisors: Prof. Hiroya Takamura, Dr. Tatsuya Ishigaki)
Aug 2021 – Aug 2022 · Tokyo, Japan
Conducted research on data-to-text generation and LLM agent collaboration for natural language processing.
Work Experience
Co-Founder & CTO
Kiwami AI
Nov 2024 – Jun 2025 · Tokyo, Japan
Engineering Intern (Full-time)
Turing Inc.
Jul 2024 – Aug 2024 · Tokyo, Japan
Machine Learning Engineer (Full-time)
Medical Dataway AG
Oct 2023 – Jul 2024 · Zurich, Switzerland
Data Analyst Intern (Part-time)
UTokyo Economic Consulting Inc.
Nov 2020 – Apr 2022 · Tokyo, Japan
Software Projects
AI-Polis
Consensus-building agent framework that coordinates independent LLM votes with feasibility checks for policy ideation.
Aug 2025 – Present · TypeScript · LangChain · LLM Agents
MultiAgent Research Ideator
Tooling that scaffolds research brainstorming sessions through scripted multi-agent dialogues.
Mar 2025 – May 2025 · Python · OpenAI API · FastAPI
Nikkei-UTEcon Daily Economic Indicator
Economic indicator pipeline producing technical reports from 40 years of news data.
Dec 2020 – Apr 2022 · Python · NLP · Time-series Analysis
Publications
Peer-reviewed
Pioneering AI in surgery: Deep learning for scene recognition in laparoscopic hysterectomy
Giovanni Ruggeri, Naoki Ito, Keisuke Ueda, Mineto Fujisawa, Daiki Hiratsuka, Takashi Matsumoto, Michael Mueller
European Journal of Obstetrics and Gynecology and Reproductive Biology, Volume 313 · 2025
Exploring the Design of Multi-Agent LLM Dialogues for Research Ideation
Keisuke Ueda, Wataru Hirota, Kosuke Takahashi, Takahiro Omi, Kosuke Arima, Tatsuya Ishigaki
Special Interest Group on Discourse and Dialogue (SIGDIAL) · 2025 · Avignon, France
Real-Time Integration of Optical Coherence Tomography Thickness Map Overlays for Enhanced Visualization in Epiretinal Membrane Surgery: A Pilot Study
Ferhat Turgut, Keisuke Ueda, Amr Saad, Tahm Spitznagel, Luca von Felten, Takashi Matsumoto, Rui Santos, Marc D. de Smet, Zoltán Zsolt Nagy, Matthias D. Becker, Gábor Márk Somfai
Biomedical Applications of Optical Coherence Tomography, Second Edition · 2025
Real-Time Optical Coherence Tomography Overlay in Epiretinal Membrane Peeling for Enhanced Intraoperative Visualization: A Pilot Study
Keisuke Ueda, Amr Saad, Luca von Felten, Takashi Matsumoto, Ferhat Turgut, Gabor M. Somfai, Matthias Becker
The Association for Research in Vision and Ophthalmology (ARVO) · 2024 · Seattle, United States
Additional
Automated Surgical Report Generation Using In-context Learning with Scene Labels from Surgical Videos
Keisuke Ueda, Mineto Fujisawa, Daiki Hiratsuka, Giovanni Ruggeri, Takashi Matsumoto, Michael Mueller
Hamlyn Symposium on Medical Robotics (Poster) · 2024 · London, United Kingdom
Note: Runner-up Poster Award
Kinematic-aware Generation and Classification of Commentary in Racing Games
Keisuke Ueda, Tatsuya Ishigaki, Ichiro Kobayashi, Yusuke Miyao, Hiroya Takamura
Preprint · 2024
Old Japanese Character Recognition by Convolutional Neural Net and Character Aspect Ratio [Japanese]
Keisuke Ueda, Motoharu Sonogashira, Masaaki Iiyama
ELCAS Journal, Volume 3 · 2018