Events

Prior to 2024, most talks were internal to Microsoft and Amazon. The list below highlights recent public-facing talks, invited workshops, and presentations.

2026

  • ICMS AI × Mathematics Workshop, International Centre for Mathematical Sciences, 23-27 February, Edinburgh, Scotland.

2025

  • GCHQ CyberFirst Programme, The Current State of AI Security, 1 August, Microsoft London, UK.
  • Applied Topology Conference, Persistent Homology for LLM Interpretability, 14–18 July, Adam Mickiewicz University, Poznań, Poland.
  • London–Oxford Topological Data Analysis Seminar Day, Holes in Latent Space: Topological Signatures of Adversarial Influence, 22 May, London, UK.
  • United Nations Cyber Norms Workshop, Enhancing Capabilities for Detecting Adverse Cyber Activities, 19–20 May, Berlin, Germany.
  • IEEE Secure and Trustworthy Machine Learning Conference (SaTML), two contributed talks (main track and competition track), 9–11 April, Copenhagen, Denmark.
  • GirlsWhoML Workshop, Interpreting LLM System Behavior with Basic Math and Statistics, February, Entrepreneur First, London, UK.
  • Zero Day Quest: Microsoft’s Bug Bounty Program and AI Research, online traning, January. video.

2024

  • AI Fundamentals for Civil Servants, "The Long View on AI: What’s Next and What Matters" panel discussion, December, Imperial College London, UK.

2023

  • Microsoft BlueHat, Breach Path Busters: Decoding Cyber Threats using Graphs and AI, Washington, USA. video.
  • United Nations Internet Governance Forum, Kyoto, Japan.
  • NATO European Cyber Conflict Research Initiative, Brussels, Belgium.

Research Visits

2025


Poster Presentations

2025

  • Applied Topology Realization Network Conference, Holes in Latent Space, poster presentation, August, Chicago, USA.
  • Erlangen AI: Mathematical Foundations of AI Conference, 9–11 June.
  • London School of Geometry and Machine Learning Summer School.
  • IEEE Secure and Trustworthy Machine Learning Conference (SaTML).

Summer Schools

  • London School of Geometry and Machine Learning Summer School,
    Accepted project: Cycle Matching for LLM Interpretability using the TaskTracker Dataset, 7–11 July, 2025.