I'm a Ph.D. student in Computer Science at the University of Chicago (since Autumn 2024) advised by Ari Holtzman. My research focuses on interpretability of neural networks. I'm part of the Communication & Intelligence group and also collaborate with Victor Veitch, Aaron Schein, and Globus Labs.

Previously, I studied Chemical & Biomolecular Engineering at University of Illinois Urbana-Champaign and worked as an environmental consultant. Then, I spent awhile in the fitness industry — I own a gym in downtown Chicago called South Loop Strength & Conditioning and an individualized coaching company called Legion Strength & Conditioning. I received my M.S. in Computer Science from the University of Chicago, then worked as a data scientist for the Data Science Institute.

Links of interest:

I also occasionally send out a newsletter. Subscribe below if you're interested!

Research

  • Sok: On finding common ground in loss landscapes using deep model merging techniques
    Arham Khan, Todd Nief, Nathaniel Hudson, Mansi Sakarvadia, Daniel Grzenda, Aswathy Ajith, Jordan Pettyjohn, Kyle Chard, Ian Foster
    arXiv (2024)
  • RATE: Causal Explainability of Reward Models with Imperfect Counterfactuals
    David Reber, Sean Richardson, Todd Nief, Cristina Garbacea, Victor Veitch
    arXiv (2024)