Constructing a multilingual benchmark for evaluating the safety of AI systems in mental health applications at Nara Institute of Science and Technology (NAIST).
Developing an enterprise knowledge platform enabling analogy-based search through knowledge graphs and multimodal retrieval at Matsuo Institute, Inc.
Developing an AI agent system to support enterprise sales activities using LangGraph, FastAPI, and React (TypeScript) at Matsuo Institute, Inc.
Developing a system to automatically evaluate the quality of business proposal documents using Vision-Language Models (VLM), as part of the University of Tokyo Matsuo Laboratory.
Contributing to the development of a medical ontology aimed at structuring free-text entries in electronic health records, using LLM at Nara Medical University.
Developing an LLM-powered psychotherapy training tool using Streamlit, prompt engineering, and RAG at Nara Medical University.
Developed data analysis competitions and business proposal assignments using Python libraries as part of the University of Tokyo Matsuo Laboratory.
Adapted the AlphaPullDown package for cloud-based workflows using Python at Waseda University Asahi Laboratory.
Developed BERT fine-tuned models for customer retention prediction and clustering using real-world data in the University of Tokyo Matsuo Laboratory GCI program.
Built a mobile app with facial recognition for stress estimation using Microsoft Azure Custom Vision API as part of the University of Tokyo 100program.