Curriculum Vitae

University of Tokyo
Research Interests: Scientific ML, Fusion Plasma Wave, Digital Twins
Advisor: Prof. Naoto Tsujii

Imperial College London
Dissertation: Physics-Informed Neural Operator Scheme for Accelerating Simulation of 2D Lower-Hybrid Wave on TST-2 Spherical Tokamak (Merit)
Advisor: Prof. Adriana Paluszny Rodriguez; co-advised by Prof. Naoto Tsujii

National Central University, Taiwan — with credit program in Interdisciplinary Artificial Intelligence.
Advisor: Dr. Chun-Yu Lin and Prof. Jih-Hong Shue

Feasibility studies, prototypes, and technical advisory for simulation acceleration.

End-to-end clinic booking and queue management system with LINE chatbot; AI-assisted clinical pipeline (ASR + summarization) on a privacy-aware, event-driven backend.

• Dynamic patient registration cutoff optimization
• Server redesign, deployment, and monitoring

Agentic transaction modelling; production ML system architecture.

Integrated neural operators (PINN, FNO) into a production simulation API: ~2000× speedup (hours → seconds) with less than 3% error; delivered a prototype neural-operator solution for plastic molding simulation.

Developed machine learning and numerical solutions for space physics.

• Space weather: magnetopause modelling with physics-based machine learning
• Remote sensing: particulate matter (PM0.3, PM2.5) interpolation with spatial random forest — enhanced spatiotemporal resolution (1 hr → 10 min, 12× sampling rate); RMSE 11.65 → 0.62 [AI Space Challenge]
• Satellite maneuver: numerical and ML/NN simulation for CubeSat ADCS

Benchmarked PINN, DeepONet, and FNO models on Xeon clusters and NVIDIA A100 DGX; modular TensorFlow implementation of FNO, including a first TensorFlow-based neural operator demo on MNIST/CIFAR-10 and a physics-informed neural network for the advection–diffusion equation.