Avatar

I am a Research Scientist at Anthropic and PhD student at MIT CSAIL (on leave) advised by Jacob Andreas and Julie Shah. I’ve spent summers at the Boston Dynamics AI Institute, MIT-IBM Watson AI Lab, Facebook AI Research (FAIR), and before grad school, two years as an AI Resident at Microsoft Research. I did my undergrad at Yale, where I got my start in research with Brian Scassellati and read a lot of dead philosophers.

I’m interested in building agents that learn representations from rich human knowledge, whether directly (e.g. from users) or through priors (e.g. from LMs). Currently, I’m thinking a lot about how to utilize pretrained models in conjunction with human feedback to interactively learn aligned preferences/rewards.

A history buff at heart, I care deeply about working with non-academic communities to create safe, ethical, and equitable AI. I currently serve as a Special Government Employee for the Defense Innovation Unit (DIU). In a previous life, I worked at the White House Office of Science and Technology Policy (OSTP), National Institute of Standards and Technology (NIST), and Schmidt Futures. I also serve on the advisory board of the Yale Jackson School of Global Affairs, where I co-teach a course on AI for policymakers.

I love being outdoors, even in the brutal Boston winters. A current goal is to run a sub-3:00 marathon (this is how I’m doing). Reach out to chat about research, policy, or running! Preferred subject line: Your cat is dope.

email | cv | google scholar | twitter | linkedin

Education

  • Ph.D. Computer Science, 2023 -

    Massachusetts Institute of Technology

  • M.S. Computer Science, 2023

    Massachusetts Institute of Technology

  • B.S. Cognitive Science, 2018

    Yale University

  • B.A. Global Affairs, 2018

    Yale University

People Financially Invested in My Future

  • Open Philanthropy
  • NSF Graduate Research Fellowship
  • Truman Scholarship
  • My parents

Recent News

All news»

[Sep 2024] Our paper Adaptive Language-Guided Abstraction from Contrastive Explanations was accepted to CoRL 2024.

[Aug 2024] I am taking leave from MIT to lead national security evaluations on the Frontier Red Team at Anthropic.

[Aug 2024] Attending RLC! I’ll be presenting Pragmatic Feature Preferences in the RL Beyond Rewards Workshop.

[Jul 2024] Attending ICML! I’ll be presenting Pragmatic Feature Preferences in the main conference. I’ll also be attending the Alignment Workshop beforehand.

[Jul 2024] Attending RSS! I’m honored to be part of the 2024 RSS Pioneers cohort, as well as help organize the Social Intelligence in Humans and Robots Workshop and the Task Specification Workshop.

[May 2024] I started at Anthropic! I’ll be working to help make big models safer.

Publications

*
Adaptive Language-Guided Abstraction from Contrastive Explanations
Constrained Human-AI: An Inclusive Embodied AI Assistance Challenge
Learning with Language-Guided State Abstractions
Aligning Robot Representations with Humans
Preference-Conditioned Language-Guided Abstractions
Getting Aligned on Representational Alignment
Human-Guided Complexity-Controlled Abstractions
Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback
Diagnosis, Feedback, Adaptation: A Human-in-the-Loop Framework for Test-Time Policy Adaptation
Strengthening Subcommunities: Towards Sustainable Growth in AI Research
Make Greenhouse-Gas Accounting Reliable — Build Interoperable Systems
Investigations of Performance and Bias in Human-AI Teamwork in Hiring
On the Nature of Bias Percolation: Assessing Multiaxial Collaboration in Human-AI Systems
Human-Machine Collaboration for Fast Land Cover Mapping
What You See Is What You Get? The Impact of Representation Criteria on Human Bias in Hiring
An Integrated Machine Learning Approach To Studying Terrorism
Conceptual Feasibility Study of the Hyperloop Vehicle for Next-Generation Transport
Early Detection of Boko Haram Attacks in Nigeria