About

I'm the founder of Guardal, where we build counter-swarm command and control: software that lets small teams defend against drone swarms numbering in the thousands.

Previously, I used RL to improve adaptive perception and foveated sensing in embodied robotic agents. My ideas included developing adaptive sensors that dynamically switch modalities based upon prediction errors and control systems that take input from these dynamic sensors. The unifying theme was that perception, decision making and physical action need to be designed together. I am also interested in deep learning theory, and I write about it from time to time in otium.

The throughline from that work to Guardal is decision-making under scarcity. Foveated sensing asks where an agent should spend limited attention; adaptive sensors ask when a measurement is worth its cost; the controller has to act on whatever the sensors could afford to see. Countering a drone swarm is the same problem under fire: too many tracks, too few sensors and effectors, and a network you cannot trust. Guardal is that idea built into software. A selection of the earlier work is on the research page.

You can reach me at .