About Me

As a professional Data Scientist, I pride myself on devising novel uses of data to generate insight. I recently received my M.S. in Analytics from Georgia Institute of Techonology after seven years consulting in workforce analytics and cloud integrations, primarily with Deloitte. I am a life-long learner and continue to take classes on computing, AI, and advanced mathematics during my spare time. I am also currently serving as an alumni advisor to the Georgia Tech Robocup team, which I had the priviledge of being a part of while pursuing my M.S.

At a personal level, I am a math and computing enthusiast, tinkerer, and general purpose nerd. I spent my childhood on the math team, received my B.S. in Applied Mathematics from Georgia Tech in 2010, and continue to enjoy discussing mathematics over a drink when a colleague or friend is willing (especially when it's focused on graph theory or machine learning).

I've been skiing since I was four, and it remains my favorite winter activity, particularly at Whistler/Blackcomb. However, having moved back to the southeast it may be some time before I'm able to return. I also play inline and ice hockey on the weekends (quite poorly) and am an avid Dallas Stars fan. You can often catch my dog, Leopold the Great (Pyrenees), and I on the couch watching hockey, while I rub his belly. You can see Leo's beautiful face pictured above, and you can also follow him on Instagram.

Blog

  1. Fantasy Hockey Draft Prep with Altair

    I recently had a need to embed plots into my blog (for an upcoming post I am working on), and my favorite python plotting library right now is definitely Altair, so I figured I would share a bit of the testing. This post is essentially a light how-to and mild evangelism for Altair. Altair is a declarative visualization library, which means you can focus on what you want your visualization to do, and not on how you want it to get done. I won’t go into specifics, but if you’re interested this is a great talk from PyCon by the main contributor, Jake VanderPlas. …


  2. NLP - Scraping a Real World Corpus

    There are quite a few text corpora readily available in the standard NLP python packages (NLTK has quite a few). However, using those corpora would gloss over the practical issues that arise from real world data, which I think are worth discussing. Instead, I am using a Kaggle dataset of Huffington Post news headlines and scraping the text of those articles to build a corpus. …