I am a machine learning researcher and Senior Scientific Associate (AI) at the University of Cambridge, based at the Cancer Research UK Cambridge Institute. My research spans neural architecture design, generative and representation learning, and AI for scientific discovery, with a particular focus on biology, cancer research and drug discovery.

At Cambridge, I work directly with experimental cancer scientists. My current research includes predicting cellular responses to genetic perturbations and virtual-cell modeling, alongside scientific AI agents that retrieve and reason over experimental evidence. I am interested in models that generalise beyond observed experiments, connect heterogeneous scientific data with biological knowledge, and help researchers investigate the effects of interventions.

My research combines original ideas with hands-on model development and experimental validation. I conceived and built DeepEventMine, a neural architecture that jointly learns interdependent tasks and recursively constructs complex biological event structures from scientific text. I developed BioVAE, adapting a variational language-modelling framework to investigate biomedical representations and their transfer across tasks. Both were published in Bioinformatics. More recently, I developed new initialization and training methods to enable effective iterative refinement in diffusion models, published at ACL 2026. Earlier, I led the development of a machine translation system that achieved runner-up in English–Vietnamese translation at IWSLT 2015.

Previously, I conducted biomedical AI research at the National Institute of Advanced Industrial Science and Technology (AIST, Tokyo, Japan) and worked as a Senior Researcher at the National Centre for Text Mining, University of Manchester. These roles shaped my research in structured prediction, biomedical information extraction and generative representation learning.

Alongside academic research, I have substantial experience developing AI for real-world use. At FPT Smart Cloud, I led AI teams and served as a lead architect for FPT AI Agents. My work included language-model fine-tuning, AI architecture and core implementation, taking banking applications through production deployment and customer acceptance for commercial use.

I received my PhD from the Japan Advanced Institute of Science and Technology in 2017, under the supervision of Professor Minh Le Nguyen, with research in natural language processing and machine translation. Throughout my career, I have also benefited from the mentorship and collaboration of Professors Jun’ichi Tsujii, Sophia Ananiadou and Makoto Miwa. I have taught deep learning and data mining and currently lead a student research group exploring generative models and AI for scientific discovery.

My research interests centre on machine learning for biology and drug discovery, particularly generative models, cellular-response prediction, and AI systems for scientific reasoning and discovery.