Google DeepMind Research Scientist
Senior Research Scientist at Google DeepMind, specialising in reward modelling, alignment, and multi-agent systems for large language models.
I am a Senior Research Scientist at Google DeepMind, where I lead research on novel approaches to reward modelling and alignment. My work focuses on large language models operating in multi-turn, long-horizon settings.
This includes exploring novel signals that emerge from LLM agents interacting with people and with one another — behavioural, structural, and dynamic properties of these interactions that can be leveraged to build better-aligned, more capable systems. I am particularly interested in the coordination and communication structures that arise in multi-agent settings, and what these reveal about how to scale alignment beyond the single-model case.
Before DeepMind, I completed my PhD in Computer Science at the University of Cambridge, specialising in machine learning for language, as a Harding Distinguished Postgraduate Scholar.