Currently living in Seattle while working on the Excel team at Microsoft, helping to develop a better charting experience. Good things to come!
Worked with neuro-linguistic programming and principles of machine learning. Found trends and patterns in open comments regarding customer experience with different companies and related sentiment to each comment using a tone analyzer API in Python. Developed interactive graphs and charts using d3 to plot the data found. Researched and analyzed multiple Sentiment Analysis APIs using scikit-learn and found optimal Sentiment classifier by combining mulitple APIs.
Worked with an agile team on a Proactive Messaging project developed in Java. Completed an automated test using JBehave and Selenium web drivers that scripted test scenarios of said project, shortening the testing time by 60%.
Datalions is a cross-platform application that helps the fitness team and coaches of the Orlando City Soccer Club predict the risk of soft-tissue injury for players using machine learning and data visualization. It also serves as an injury log system for the team. The application was sponsored and is currently being used by the Orlando City SC.
Outfluenza is a React Progressive Web Application that leverages four separate data sources for flu information including the Center for Disease Control and Prevention (CDC), Google Trends, Twitter, and a query for local doctors. Outfluenza focuses on small communities where news sources are not enough to determine the severity of the disease in their area with certainty. For this project, we aimed for minimalism and crowdsourcing.
Amazon's Alexa skill prompts the user how his/her day is going, then it curates a list of regularly-updated Spotify playlists tailored to the person's mood. Worked on incorporating the AlchemyAPI to the response from Alexa, then sorting a list of playlist in accordance to the sentiment in the person's day.
Receives an article (or any body of text) and produces a summary around what sentences have the most value. Reduces big bodies of text to a 70-85% of it's original length. On the server side, the program is written in Python using the NLTK library, it's made into a RESTful API using Flask and deployed to AWS Elastic Beanstalk. On the client side, the application is written using Angular and deployed to GitHub Pages using angular-cli-ghpages. Project inspired by the auto TL;DR bot used in many subreddits.
Website scrapes news articles from multiple reliable prominent sources and categorizes such articles by mood/tone of the information. I developed the Watson® Tone Analyzer processor for the articles, a web scrapper for BBC news, and the website’s CSS using Materialize. Processor and web scrapper coded in Python.
App eases the complexity of a web process to submit reimbursement forms for State Farm® employees. Developed with Ionic and AngularJS (Compatible with iOS and Android). I worked on the code for the ionic framework, and the basis for the AngularJS back end.
If you're interested in Data Analytics, you will find learning about Natural Language Processing very useful. For this article, we'll use the NLTK library suite to build a text summarizer in Python.
Since 2007, MLS has been increasing its popularity -- bringing more revenue to the organization and an opportunity to increase the budget for each team. Here is a quick analysis in the salary growth of players in the last decade.
I am a Computer Scientist from the University of Central Florida. In my free time, I like to go for hikes, play soccer, stream, and lots of 🌮 tasting.