Chirag Nagpal

AI Research Scientist,
Meta Platforms Inc.

chiragn at alumni dot cmu dot edu
[vita] [gscholar] [twitter] [calendar]







I am an Artificial Intelligence researcher interested in Safety, Reinforcement Learning and Survival Analysis. Most recently, I was a member of Meta's Superintelligence Lab, where I architected Meta's first alignment recipe for Safety with Reinforcement Learning. Previously, I was a full-time research scientist at Google Research where I worked on imitation learning and modelling human preferences. I have served as a core contributor to multiple foundation models, including Gemini, Muse Spark and Image and Llama 4.

I received my PhD at the School of Computer Science at Carnegie Mellon University, where I worked on approximate inference techniques for semi-parametric graphical models to estimate censored time-to-event outcomes and counterfactuals. I am was part of the Auton Lab and advised by Prof. Artur Dubrawski. My thesis research has seen application in multiple areas of clinical risk prediction, as well as in predicting disruption for safe control of Nuclear Fusion reactors. At Carnegie Mellon, I served on the Dean's PhD Students Advisory Committee and co-chaired the SCS Dec/5.

During my PhD, I worked as a Summer Associate in Prof. Manuela Veloso's AI Research Team within the Corporate and Investment Banking division of J.P. Morgan. I also spent a summer working with Kush R. Varshney and others at IBM Research's Thomas J. Watson Research Centre as a Science for Social Good Fellow.


Preprints

Publications

Foundation Model Contributions

Software

Workshop Presentations

Clinical Abstracts

Professional Service

Teaching

Personal

I'm a dormant member of the W3VC, the Carnegie Tech Radio Club; In my past life, I used to make (hack?) stuff.

I enjoy Equitation, playing the Guitar, and Trivia and Quiz contests.

Here's a [list] of all the places I lived in before coming to Pittsburgh.

[phd thesis]
***