Research

My current work at:

  • Meta (formerly Facebook) focuses on developing systems for detecting inauthentic behavior on our platforms.
  • University of Wisconsin-Madison focuses on using deep learning models to predict bedrock depth from aerial lidar images.

Graduate Research

My graduate research focused on two main areas:

  1. Modeling and uncertainty quantification of in-situ combustion chemical reaction models
  2. Imaging methods for source rock characterization

The unifying theme between these areas is the application of statistics, machine learning, imaging, and optimization to develop data-driven methods in laboratory-scale study of enhanced oil recovery. Click on the links above to learn more about each line of work.

Teaching

While at Stanford, I was very active with the ACE Program, a supplementary instruction program targeted towards students from underresourced backgrounds to create greater equity and opportunity in the School of Engineering. My work with ACE included:

  • Course assistant for CME 100 and CME 102 for a total of 7 quarters
  • Program coordinator during the 2018-2019 school year, helping to establish new policies and procedures, managing enrollment and the teaching staff, and substitute teaching when needed.

As a course assistant, I created many resources for my students that I’ve made available online:

Please visit my Resources page to view materials I wrote for my ACE sections and other academic resources, and my About Me for further details about my teaching activities.

Public Policy

I am interested in public policy related to the regulation of technology, specifically data ownership and privacy, social media, and AI technologies in the context of energy systems. In the summer of 2019, I participated in the Hoover Institution Summer Policy Bootcamp. As part of this bootcamp, I wrote a policy proposal titled “Less Can Be More When Regulating Deepfakes,” for which I won the Hoover Institution Director’s Award.

Internship Projects

I have interned at Intel (June 2017-December 2017) and Facebook (June 2020-September 2020) during grad school. At both companies, I had the opportunity to explore some very interesting projects related to AI systems and deep learning:

  1. Intel: applications of higher-dimensional algebraic systems for distributed training of vision models using forward-mode autodifferentiation.
  2. Facebook: deep generative models for synthesizing recommendation system datasets, with specific applications to Neural Collaborative Filtering (NCF) and Deep Learning Recommendation Model (DLRM).