Hi, I'm Tyler Kaye!

Interning at MongoDB gave me a great introduction to systems programming. As a member of the Platforms Server team, I contributed to many of the core processes involved in the machinery of the core MongoDB server. This experience not only enhanced my programming abilities, but also it exposed me to many important software engineering principles and practices.

Main Project: Tasked with beginning the MongoDB Mobile Project to create an embedded database capable of running on a smartphone. This involved creating and exposing a C-API that connected the server and the client through a minimal transport layer. Then, the MongoDB C-Driver was connected on top of the custom C-API and cross-compiled in order to run on an IPhone.
Main Project #2: Began the development of a Swift Driver for MongoDB that wraps the C-Driver and enables easy interaction with the database from Swift. This work involved a lot of research into the backwards compatibility of Swift and how to properly attach Swift code to legacy C-API's.
Other Work: Worked on a variety of tasks throughout the course of the summer ranging from bug fixes to feature addition. Two main patches involved increasing the efficiency of the BSON extractor for the database and re-writing the URI parser in order to fully adhere to the specifications.

Technologies Used: C++, C, Swift, AWS, Git, Scons

MasterCard provided me a unique opportunity to learn how the payments industry works. While I was a software architecture intern, I was exposed to and worked with cutting-edge digital payment processes (most notably MasterPass). Beyond the programming though, I learned about the complex relationships in the payments industry with both banks and consumers. Furthermore, I was able to work on several different projects that exposed me to an array of issues and complications that confront the digital payments industry.

Project #1: Worked on the SDK generator and its implementation into the MasterCard developer portal for its OpenAPI initiative.
Project #2: Helped to develop the POC for an iOS embedded registration framework using state-of-the-art security protocols.
Project #3: Tasked with looking into the systems/products of other competitors and provided a recommendation on potential changes / limitations to the platform.
Project #4: As a part of the Intern Business Challenge, a team of other interns and I developed MasterCrop: a SMS and web-based platform to remove the need for middlemen in the agricultural industries of third world countries by pairing merchants with local farmers and handling the transaction over MasterCard’s payment rails. We received second place of 8 teams in the competition.

RipeOrWrong is an Android application that identifies bruised and under/over-ripened fruits and vegetables. The application utilizes the FLIR One thermal imaging camera in order to measure thermal variations below the skin of the fruit / vegetable. Additionally, the application employs a deep convolutional neural network designed with the TensorFlow framework in order to implement the classifier. Although this application focusses on apples, it has been designed and documented so that the implementation of other fruits and vegetables is seamless. RipeOrWrong leverages both thermal imaging and modern machine learning to prevent consumers from purchasing undesirable fruits and vegetables

Major Components:

1. Developed seam carving algorithm to generate a robust dataset of thermal images of fruits
2. Trained two different convolutional neural networks to grade and identify poor quality fruits and vegetables
3. Integrated the more successful network into an Android application utilizing the FlirOne thermal imaging camera SDK

Wager is a social media gambling application that enables peer-to-peer gambling. The application uses GeoFencing techniques to encourage users to make bets with those around them in their community.

My Major Contributions:

1. In charge of the back end of the application including integrating Firebase into the backend
2. Currently in the process of migrating the database of the application to MongoDB Stitch
3. Integrated GeoFire to power the geolocation-based features of the application
4. Incorporated Venmo API into the application to facilitate easy peer-to-peer payments