Hello! I am a PhD candidate in the Department of Mathematics at the University of Washington in Seattle, where I am lucky to be co-advised by Dmitriy Drusvyatskiy and Rekha Thomas. My research interests broadly consist of questions inspired by data science and machine learning, with a particular focus on problems with an algebraic or combinatorial nature. More specifically, my recent work has concerned kernel methods.
Prior to my time at the University of Washington, I was an undergraduate student at Smith College in Northampton, MA.
Outside of math, I can be found rowing and coxing on Lake Union, watching movies, and reading both fiction and nonfiction.
As of Fall 2025, I am on the job market.
This project was completed as a part of the Math 2 Power Industry workshop in 2024.
Illegal and unreported fishing is a global problem that is estimated to contribute a loss of up to 26 million tonnes of fish annually, yet there is currently no extensive database of incidents worldwide. We proposed a dynamic system for the automatic generation of search engine prompts that yield reports of incidents related to illegal and unreported fishing. We also proposed two metrics to evaluate the relevancy of a prompt’s results to the topic at hand. By implementing an automated system of prompt generation and evaluation, large scale searches of the internet can be conducted autonomously, leading to the acquisition of larger amounts of relevant reports than would be possible with manual effort.
This project was completed as a part of the Erdos Institute’ Data Science Bootcamp in Spring 2024.
Peer reviewed articles are a foundational pillar of academia. In theory, the process of peer review ensures that high quality, credible, and trustworthy research that advances our current knowledge can be disseminated to the community at large. However, it is not a foolproof process and some papers do fall through the cracks and end up being retracted. Retractions indicate seriously flawed and unreliable research, errors, fraud, ethical issues, or other serious concerns. Our aim was to build a classifier that identifies papers that have a high risk of retraction. We hope that our model could be helpful in the peer-review and publication process but want to emphasize that it cannot and should not replace rigorous scrutiny from an expert.
My email is jackgk (at) uw (dot) edu.