I'm a second-year PhD student at MIT CSAIL, advised by Dylan Hadfield-Menell. My research centers on aligning AI systems — such as LLMs — with human goals.

I'm especially interested in problems of

  1. Scalable oversight and aligning superhuman AI systems
  2. Alignment to humans with diverse preferences
  3. Making LLM pre-training safe and data-efficient
I like building methods that leverage clear theoretical insights on important problems. I'm also interested in the intersection of AI and economics, and how tools and frameworks from economics can be used to achieve technical and social goals in AI alignment.

Previously, I obtained my bachelor's degree at Johns Hopkins, where I worked with Professor Rene Vidal on the theory of deep learning. During undergrad, I also worked with Dyno Therapeutics and NASA on ML-guided protein design.

Outside of work, I enjoy playing piano, backpacking, salsa dancing, and philosophy.

Please reach out at stew@csail.mit.edu if you want to talk!

Google Scholar Resume Goodreads