Click the links on the left to learn more.
I'm a rising senior at the University of Texas at Austin, seeking degrees in computer science, mathematics, and government. Since May 2014, I have been a member of the Turing Scholars, the computer science honors program at UT. I am a pod mentor, responsible for mentoring and advising new students, introducing them to the facilities and programs available to them, and helping them with the transition between high school and college. I am also a Computer Science Ambassador; I give tours of the building and chat with prospective students about being a student at UT.
I am currently working on my Turing honors thesis, for which I am investigating how gerrymandering might be solved with approaches in artificial intelligence. I am planning on defending my thesis and graduating in December 2017.
I was recently named a 2016 Archer Fellow. The Archer Fellowship is a semester-long program where students from the UT system spend a semester in Washington, DC, working full-time at an internship in the city while taking courses taught by UT professors. Around 50 students are selected each year; I am one of two computer scientists in my year and the only mathematics major to have been selected to the program.
My Archer internship was at the White House Office of Science and Technology Policy, where I served as a staffer intern for the President's Council of Advisors on Science and Technology.
I wrote for the department blog for two and a half years; some of my articles have received more than a thousand reads, and more than one student has informed me that my article on my first year at UT was the deciding factor in their decision to come to Austin. Furthermore, in Spring 2015, I had an op-ed published in the Daily Texan, UT's school newspaper; read the article here.
I worked for three summers in high school at different labs, writing code and helping conduct research. I spent a summer building the tech stack at a small real estate management company in Houston, including setting up servers for data management. I interned at Bloomberg in New York City from May until August 2014, where I worked on building a visualization tool for our volatilities data. I interned at Apple Inc. in Cupertino, CA from May until August 2015, where I worked on the Proactive Assistant team building internal tools and prototyping new software. I interned at Palantir in Washington, DC during summer 2016.
I am also a singer, and since January 2015 have served as the Director of the Texas Songhorns, helping them learn new music and setting the organization's goals. (I am also our webmaster.) In December 2014, I was asked by my alma mater, St. John's School in Houston, to perform as the soloist at our Christmas mass, and sang Ralph Vaughan Williams' Fantasia on Christmas Carols, a personal favorite of mine. I also play the piano, although much less diligently.
AnalyticsBot is a bot that runs on top of GroupMe, receiving commands and processing them. While its main feature is user analysis -- number of likes given to or by a user, most commonly used words, etc. -- it also has a rudimentary imitation engine. The engine takes in as a corpus all of the user's messages, sorted in weight by number of likes, and uses a Markov chain to generate the simulation. GitHub
Developed for HackDFW 2015, foodme is an app that connects users with people in the area interested in getting food at local restaurants. foodme uses your location to find restaurants near you, lets you select the date and time you want to get food, the number of people you want to get food with, and adds you to a group; when your group is filled, everyone in the group gets a notification and is added to a group chat (powered by GroupMe). GitHub
πPrint (pronounced "pi-print") is a tool that sets up a CUPS-based print server on a Raspberry Pi connected to a printer. A website allows users to upload files onto the RPi (via a PHP script that runs some basic sanity checks against malicious users); a bash script is loaded every minute via a cronjob that finds connected printers, loads the first one, and prints everything in the upload folder. The uploaded files are then archived, and eventually deleted for security reasons. GitHub
L++ is an extension of the toy functional programming language L. Developed with Arnav Sastry and Tres Popp for our final project in our Programming Languages class, L++ adds type inference, list polymorphism, type inference, and sockets to L. The infile is lexed using flex, parsed using bison, and then evaluated in a suite written in C++.
One of the projects assigned in honors Distributed Computing was to implement a chat room, backed by the asynchronous consensus algorithm Paxos. Paxos tolerates all but one server crashing and our implementation further allows for servers, once crashed, to come back to life and rejoin the system. The project was implemented over the course of two weeks in Java along with Kyle DeHolton.
Developed for HackRice 2014, mo (modulus operandi) is a service that provides secure end-to-end encryption for file uploads. mo's primary feature is that all data is encrypted by the client, as opposed to by the server, to protect user privacy from man-in-the-middle attacks. Files can be uploaded via a command-line script, and can be shared in small groups with a known password.