I am a Research Engineer in the Alignment Research Group at NYU's Center for Data Science. Previously, I was a Software Engineer at Amazon, where I worked on improving the performance and extensibility of the core infrastructure in the advertising division. I hold a Bachelor's degree in computer science from Princeton University, where my thesis was on fine-tuning language models by applying reinforcement learning to text-based games.

I also competed in debate, where I was the runner-up at the World Universities Debating Championship, top individual debater at the collegiate North American Debating Championship, and two-time high school national champion.

Outside of work, I spend a lot of time making overly elaborate trivia questions, which you should check out here. Some older personal projects can be found here.

Publications

Training Language Models to Win Debates with Self-Play Improves Judge Accuracy
Samuel Arnesen, David Rein, Julian Michael
[arxiv] [pdf][code] [media]