Last summer, I worked within Amazon Product Search, on the Digital Relevance Ranking team. My intern project was to universalize Kindle relevance ranking models.
In my senior year, I started a new project at Professor Deb Roy's Laboratory for Social Machines (LSM). Using Target food data, I analyzed co-occurrences of food purchases.
While advanced VR headset technologies using IR sensors are quite precise, we wanted to see if we could accurately estimate gaze based simply off of an image of the eyeball. Last summer, I gave a talk at the European Conference on Eye Movements (ECEM).
Working on the Google Cardboard SDK team within Google Daydream, I created an over-the-air firmware update process to allow the Daydream controller to update its firmware from information sent by a Daydream-enabled phone.
I really enjoyed 6.141, Robotics Science and Systems. Working with my team, at the end of the semester we raced our robot through the tunnels of MIT.
As a software engineering intern, I worked on gamification of tasks important to individuals with autism, such as maintaining eye contact and understanding emotions. Technologies included OpenCV and Java.
App to save special moments for a rainy day. After winning an honorable mention at Greylock Hackfest 2015, my roommates and I published our app in the app store. (4.5 star rating, 1000-5000 installs)
Working in the Laboratory for Social Machines, my postdoctoral mentor Dr. Soroush Vosoughi and I used temporal and linguistic models to link user profiles across different social media platforms. Published in SocInfo'15.
To assist Wi-Fi tests, I created an analytics website to help test engineers quickly evaluate the state of their machines.
For our mobile development class 21W.789, we created an Android app that allows users to create and go on scavenger hunts.
As part of an iOS game development competition, I worked with a friend to create a 2-player make-3 game (think Bejeweled or Candy Crush). Make combos to send your opponent an unexpected twist!
Over my freshman year summer, I explored many classifiers and features to predict the sentiment of tweets. Ultimately, we found that context (time of day, location) proved quite valuable. Published in EMNLP.
My roommate and I created a trash-to-treasure website, where users could post their unwanted items and others could claim them.
(G = graduate, H = header/ advanced, U = undergraduate)
In my undergraduate and M.Eng. career at MIT, I've taken a lot of classes. Below are some of my favorite.
(TA = teaching assistant, LA = lab assistant, O = other)
Throughout the years, I've also enjoyed serving as a course assistant for many classes.
Fall 2017, Spring 2017
IAP (January) 2017
Spring 2015 - Fall 2015
Fall 2014 - Spring 2015