I'm a research fellow in the Stanford University Department of Aeronautics and Astronautics. I completed my PhD in Aerospace Engineering at Stanford, funded by the National Defense Science and Engineering Graduate Fellowship. My current research is focused on decision-making in safety-critical systems where operational decisions must be made made quickly and correctly in complex environments while still being explainable and understandable by human stakeholders. I love space exploration and solving hard problems with good people.

I'm currently the Executive Director of the Stanford Center for AI Safety, and a post-doctoral researcher with appointments in Mineral-X and the Stanford Intelligent Systems Laboratory (SISL). I'm currently funded by the Schmidt Sciences Foundation and Torc Robotics. My current research focuses on failure detection and safety validation of complex, high-dimensional AI systems by applying reinforcement learning to efficiently find failures in a sample-efficient manner. This work is being applied to frontier generative AI models and autonomous vehicles.

Prior to this, I started and led the Spacecraft Operations Group at Capella Space, the first US commercial synthetic aperture radar Earth imaging constellation. There I developed the first fully-automated mission operations software, realizing lights-out tasking-to-delivery of radar satellite data for a commercial constellation. I subsequently started and led the Constellation Operations and Space Safety Groups at Project Kuiper. Most recently, I was a Principal Applied Scientist at Amazon Web Services, where I worked on building software services for large-scale distributed edge compute applications.

If you're looking to get in touch, please reach out to [first name]@argoinnovations.com.

Selected Talks

Agents of Tech: Can we Trust AI

In this episode of Agents of Tech, Stephen Horn and Autria Godfrey explore the rapidly evolving world of Artificial Intelligence and ask the pressing questions: Can we trust AI? Is it safe?

Hundred Year Podcast: Improving AI Safety

In this episode, I spoke with Adario Strange to explain why the commercialization of space will continue to fuel our explorations into the Moon and Mars, and how AI-powered robots may be the primary method for deep space exploration in the future.

Thesis Defense: Task Planning for Earth Observing Satellite Systems

This is the video of my PhD thesis defense at Stanford University. I discuss the task planning for Earth observing satellite systems, past approaches, the use of Markov decision processes, and the development of a new algorithm based on the maximum independent set problem.