Ruixin Song

I work on representation learning, graph learning, and physics-informed AI.

I am currently a product engineer and founding team member at GradientX Technology, building FinTorch — an AI financial copilot centred on retrieval-augmented, multi-agent pipelines for personal financial decision support and simulation.

Before that I spent two years as a research assistant with the AISViz and MAPS labs at Dalhousie University, working on spatiotemporal trajectory representation learning and on the data infrastructure behind fifteen years of maritime shipping records. I hold an M.Sc. in Computer Science from Memorial University of Newfoundland, where my thesis used gravity-informed deep learning to forecast marine traffic for invasive-species risk assessment.

My research has appeared in Scientific Reports, Biological Invasions, and IEEE BigData. You can find the full list on the publications page, or download my CV.