Interests

The systems I enjoy most, in and out of research, are noisy and adaptive: markets, start-ups, organizations, language, football and communities. In each the task is the same, to separate signal from noise.

• Markets

  Machine-learning trading strategies, built, backtested and deployed live. FedAnalysis treats every Federal Reserve word and gesture as data.

• Venture ecosystems

  Which “AI agent” start-ups are a new paradigm, and which are the generative-AI wave under a new name.

• Organizations

  Fieldwork inside algorithmic-trading desks: who overrides the model, when and why. The subject of my dissertation.

• Language

  What persuades when the persuader is human rather than AI, and what a central banker's phrasing reveals before the market has finished parsing it.

• Football

  BallSignal turns the 2026 World Cup into a prediction problem.

• Communities

  Growing LLM+ from five people to more than 7,000, and learning which formats make a research community compound rather than churn.

Off Duty

I read literature aloud for Girton College's Ridding Reading Prize: Asimov, Le Guin and Zephaniah.