This is a collection of some of the projects I’ve worked on during my time in university. If I’m not studying or hanging out with friends, I dive head first into something new. These projects often spark new ideas and interests for me to pursue.
Chopstix is a web based application that provides teams the ability to prioritize projects and features using the RICE prioritization framework. In small companies, deciding which new features will create the most value is not an easy task. With limited data, resources and time, leveraging your teams common knowledge with the RICE framework can be an effective way to ensure you are are building value into your product. In addition, large companies have many stakeholders which are battling to get their desired features into the roadmap. Using the Chopstix platform, product managers, engineering managers and teams can come to consensus on what gets planned in.View Site
During my undergrad at UW, I worked with two other classmates on an open-ended machine intelligence project for an elective course. The project that we decided to undertake was to evaluate 7 different machine learning methods in their ability to classify a fetus into one of seven morphological patterns from cardiotocography data. The seven methods used were Support Vector Machines, Naïve Bayes Classifier, Gradient Boosting Machine, Random Forest, Logistic Regression, K-Nearest Neighbors and Multi-Layer Perceptron. The results obtained provided better accuracy in classification than those of other researchers who solved the classification problem using the same UCI data set.Read Paper
Valkyrie Labs was a concept imaging tool and software suite that will allow users of any experience level to generate 3D models of indoor environments. This was designed to be an easy utility to use yet extremely modular and powerful in terms of its end user functionality. Though the concept was never realized, some of the software was licensed to a robotics company.View Site
Bubble is a concept social media app. This app allows you to share content in unique dynamic ways by combining factors of current popular social media. People can post pictures in groups and reply to the pictures with additional pictures. Development for this concept stopped once I realized that the social media market is very saturated.
This project was created as a final project in the Systems Design Introduction to Design (SYDE 161). In this course we learned about the techniques and processes to creating an effective design to a given problem. The project was completed throughout the term. The idea was to solve a problem for a third world country that would enhance transportation, health, or resources. Our solution was designed to reduce the negative physical health effects caused by transporting water, as well as the time spent in transport. Our solution was a design that reduced the time spent fetching water each day by women and children by 75% to allow more time for other tasks, and alleviated users from carrying heavy objects above the ground. Throughout this project a low-fidelity, medium-fidelity, and high-fidelity prototypes were created. To justify the groups choice of final design, a final report was created which included many justifications such as; a situation of concern (SOC), Quality Function Deployment Chart (QFD), Functions-Means Chart, Morphological chart, and function breakdown analysis.
Snaptrax was created at the NHacks hackathon in London Ontario. The site takes an image URL and analyze the contents of the image using artificial intelligence image processing to match it with a song. The idea came from the many memories that I can associate with specific songs. I wanted a way to match my photos with a new song. Other applications that our team had discussed if the project were to continue would include matching music with videos or presentations for commercial use. A few features that didn't get finished include: A feed of recent photo's that have been snaptracked by users, A means to upload and store photos from users, and A more complex and accurate matching process. Note this is no longer supported thought an offline version can be viewed by clicking below.View