Kyle Sandell

This project was a plugin written for vim in vimscript and python. The idea behind this plugin is to pull simple algorithms that exist on the internet into your code without having to leave the editor.

Runner's Beat is an application that I wrote with a team at a hackathon. Runner's Beat tracks your current running pace and plays music at a tempo that most matches your pace. This project was built using the EchoNest API, MyoKit, and the Spotify API.

This project aimed to apply Generative Adversarial Neural Networks to catch card counters. The result was a program that runs endless games of blackjack where card counters learn to avoid being caught, and the catcher learns to identify increasingly sophisticated card counters.

Each network was trained with one of the following card counting strategies: Hi-Lo, K-O, Hi-Opt 1, Hi-Opt 2, Red Seven, Omega 2, Halves, and Zen. Along with these were control players trained to use basic… Show more

This project aimed to apply Generative Adversarial Neural Networks to catch card counters. The result was a program that runs endless games of blackjack where card counters learn to avoid being caught, and the catcher learns to identify increasingly sophisticated card counters.

Each network was trained with one of the following card counting strategies: Hi-Lo, K-O, Hi-Opt 1, Hi-Opt 2, Red Seven, Omega 2, Halves, and Zen. Along with these were control players trained to use basic blackjack strategy.

For this project I implemented my own neural network and training in an effort to gain a deeper understanding behind the principles of neural networks as well as Generative Adversarial Neural Networks. I hope to come back and visit this project in a few months when I have time, and potentially rewrite the neural network in C++ to execute faster, as well as add a graphical interface to watch the games as they happen. Show less

Memebrew was built as a website that would learn the kind of memes a user likes, in order to show them more of what they liked and personalize the experience.

Summarizer is a python implementation of the summarization algorithm used in my app, Summaryzer. The algorithm was selected as one Algorithmia's top three algorithms for the month of September 2015. The algorithm is also available on GitHub.

Summaryzer is an application that will take a lecture, a photograph of text or a textbook, or a document, and generate a summary of it, as well as pick out key terms.
Summaryzer is an application for iOS that takes a photo of text, or uses a pasted document and generates an intelligent summary of the document, and picks out the key terms as well.