About Me

  • Generative Models
  • Language as a Knowledge Representation
  • Program Synthesis
  • Compositionality
  • Emergence and Self-Organization
  • Reinforcement Learning
  • MS in Computer Science, 2018
    The University of Chicago
  • BA in Mathematics, 2018
    The University of Chicago

Update: In the Fall, I will be starting a PhD at the NYU Center for Data Science, where I will be advised by Mengye Ren!

Hi! I am a software engineer at Charles River Analytics and Visiting Student at the TTIC, where I am advised by Professor Bradly Stadie. I am broadly interested in building machines with human-like reasoning ability. To do so, I think it is important to understand the role language and communication play in human reasoning, and to explore how they can be used to build general purpose machine intelligence. I am currently investigating how language can be leveraged as abstraction to encode complex goals, actions, and tasks (particularly in RL and program synthesis).

My work at CRA has touched on Monte-Carlo Methods, probabilistic programming, NLP, and full-stack development. Outside of industry, I split my time as an open-source contributer to Eleuther AI and the Co-Chair of the Interpretability and Visualization Working Group in the HuggingFace BigScience program. At HuggingFace, our group is studying the training dynamics and emergent properties of Large Language Models, and we are always looking for new contributors. At Eleuther, I am leading a project on building natural language rationales for incremental code changes. Check out our ongoing work in the #contrastive channel on the Eleuther Discord, and feel free to join in.

Take a look at some of my projects here!