Bio


I'm a software engineer from industry taking courses at Stanford. I completed the Graduate Certificate in AI and am currently enrolled in CS329Z.

My interests are in AI, particularly evaluating AI systems and AI safety: how to measure whether a system is reliable, especially in settings where errors cause real harm. I'm currently building an open-source harness for evaluating speech systems on clinically meaningful errors.

I'm looking for collaborators on evaluation and safety research. Reach me at sguna@stanford.edu.