About Me

Hi there! I’m Qianshuo Ye (叶千硕), also known as Harry. I’m an MPhil student in Advanced Computer Science at Queens’ College, University of Cambridge, working with Prof. Mateja Jamnik on concept-based interpretability.

Previously, I studied Information and Computer Engineering at Selwyn College, University of Cambridge, earning a Master of Engineering (MEng) and a Bachelor of Arts (BA). For my MEng project, I worked with Prof. Bill Byrne at the Machine Intelligence Laboratory in Cambridge’s Department of Engineering, investigating implicit visual grounding in vision-language models for chart question answering. In summer 2026, I worked with Prof. Difan Zou at the School of Computing and Data Science, University of Hong Kong, exploring applications of mechanistic interpretability.

Research Interests

I’m interested in making AI systems more interpretable, both by understanding how models work internally and by designing models and representations that are intrinsically interpretable. In particular, I’m drawn to concept-based interpretability and mechanistic interpretability. I’m also fascinated by computational neuroscience, especially how insights from the brain can inspire new architectures for learning and memory.

Beyond Research

Outside of research, music is a big part of my life. My tastes span classical, jazz, and rock, and I play the cello, bass and guitar. I also enjoy running, bouldering, hiking, and trying out new recipes in the kitchen :)


Always happy to chat about interpretability, swap music recommendations, or hear about a recipe worth trying!