Distributed computing, scalable machine learning, and generally using data to craft something meaningful is what I spend most of my time on
I like cats and jellyfish
I worked with my friend Pablo Paredes on a project to detect stress from bio signals. Here is our work in progress.
Here is the final project paper we wrote for Stat 151b: Modern Applied Statistics and Machine Learning. We surveyed the performance of different classification techniques on bibliographic data.
Here is my bitbucket repository, which contains code for a web scraping framework and other various tools, among them a bibtex parser. This codebase was produced as part of my employment during Summer 2010 for Prof. Jim Pitman and Bibkn.
I completed many interesting projects for the Artificial Intelligence course. I implemented A* Search, Mini/Expectimax Agents, subjected Pac-Man to reinforcement learning and Bayes Nets approximations, and built a classifier for pixelated digits. More detailed project descriptions can be found at http://inst.eecs.berkeley.edu/~cs188/fa09/assignments.html
Before that, I worked at Cerebellum Capital for 2 summers and off and on during the year. They really opened my eyes to the applications of statistics and computer science, and fostered a curiosity and enthusiasm that remains today.