Adrià Garriga-Alonso

I’m an AI researcher who loves statistics and recurrent networks. For the past few years I worked on mechanistic interpretability and AI alignment: understanding what happens inside neural networks, and making AI systems do what we actually want. My best-known work is Automated Circuit Discovery.

Now I’m building Dokimasia, tools to help people realize their values when using computers. Before that I was at FAR AI and Redwood Research, and before that I did a PhD on Bayesian neural networks at Cambridge.

I grew up in Barcelona. I like wearing white pants and colorful shirts with floral patterns.