I build things that talk back.

I co-founded Rasa in 2016 to build open-source infrastructure for what the industry now calls agentic AI — because the closed, proprietary approach wasn't going to survive contact with real enterprise requirements.

The bet was that conversational interfaces would eventually matter to enterprise software, that doing them well would require real machine learning (not just keyword matching), and that the right layer to build at was open source. A decade later, Rasa's stack sits underneath a surprising number of the phone trees, support bots, and enterprise AI agents at banks, telcos, and governments.

What I actually do

Most of my time is spent on product and architecture decisions at Rasa — the kind of medium-term bets that don't show up in a sprint but shape what the platform can do in two years. The rest is writing (archive), speaking (talks), and occasional advising.

Recurring theme: agents that don't fall over the first time a real user touches them.

Before Rasa

PhD in machine learning at the University of Cambridge. Undergrad in Chemical Physics at the University of Edinburgh — far enough back that the path from materials-science simulations to dialogue transformers feels less strange to me than it looks on paper.