I am a computer scientist from St. Louis. I started off making games in Multimedia Fusion when I was 10, before learning "real" programming the next year, and shortly later began developing software utilities used by thousands of players of the game Pardus. I did contract work for LetsGet.Net (now the Online Ordering division of Granbury Restaurant Solutions) throughout high school, and entered Carnegie Mellon University in 2009 to concurrently work on a computer science/mathematics double major and mathematics masters.
One day, in early May of 2010, I woke up and something snapped. I had seen too many beautiful and ambitious software projects grow, and change into a hideous beast that would need to either be chained and fed for years, or killed and remade. Something was wrong -- we as a society just did not know how to engineer software. In particular, I noticed that making architectural changes was conceptually easy, but could easily require man-years of effort (and, in the case of Y2K, inspire many TV shows about the collapse of society). I began thinking about how to close the gap, and took courses in areas that would help -- type theory, formal logic, compilers, and verification -- and later become an undergraduate researcher in the Institute for Software Research under Jonathan Aldrich.
In 2012, I was named a 20 Under 20 Thiel Fellow, and left school to work on automated software reengineering full-time. I dropped out of the masters program, but spent a month of summer classes to complete a dual-degree in computer science and mathematics, having taken too many courses in my double major.
While studying the restructuring of high-level source code, I founded Project Ironfist, a mod for Heroes of Might and Magic II, experimenting with binary reengineering. We've successfully deployed novel techniques for modifying the game's internal structure, enabling major modifications at a fraction of the cost of traditional techniques.
What else? Outside of program analysis and related fields, my academic interests are largely in machine learning and statistics. My current favorite proof is of Doob's Forward Convergence Theorem, which can be paraphrased as "This supermartingale converges, because if it didn't, you could make a lot of money." Before becoming a researcher, I worked as an engineering intern at Facebook and Two Sigma Investments, and as a TA for Luis von Ahn. I studied game design under Jesse Schell. I was a member of CMU's Putnam team in 2010, and competed at the ACM-ICPC World Finals in 2012. I received a 2nd-degree black belt in Tae Kwon Do in 2009, and currently practice Shotokan Karate. I am actively involved in the LessWrong community. I also enjoy playing piano and Super Smash Bros. Melee.