Patrick O’Neill

I’m interested in applying computational, physical, and biological methods to problems in AI alignment. I’m particularly interested in the questions of how complex systems can admit simple effective descriptions.

Previously I was a founding ML engineer at Suno, and over the course of my career I’ve worked on a variety of problems in drug discovery, genomics and finance. Before that, I completed a Ph.D. in Biological Sciences at UMBC, working with Ivan Erill on information-theoretic models of transcriptional regulation.

More generally, I’m interested in philosophy, computational statistics, intellectual history, institutional design and applied mathematics. I enjoy singing Renaissance polyphony, and my Erdős number is, undeservedly, 3.

Recent Writing