I had the initiative to take a half-completed mobile app, convert it to a functional solution with ASP.NET Web Forms, then finally rewrite it as a fully operational ASP.NET MVC web application that was built using test-driven development and clean architecture. This app is now the main web application for all customers of Imalogix. Since this project's inception, I have been the most senior developer in charge of leading it, as there were no current full-time employees that ever worked with this codebase. It was just me and the 2 other previous Drexel coops hired alongside me. Additionally, as this is a small company, I have had the privilege to gain the respect of CEO John Heil, discussing requirements and ideas with him regularly. Throughout my time at Imalogix, he has encouraged me to do what I think is right, and allowed me to make the best product I can for Imalogix.

I have had the great opportunity of gaining experience in so many areas of software development. I discovered ASP.NET MVC and was able to exercise my newly acquired love for test-driven development every day since I was hired in 2019. I learned about Clean Architecture to organize the codebase, and setup the Autofac Dependency Injection framework to facilitate inversion of control. I found the Dapper ORM to make data services communicate with SQL easier, worked with a consultant to implement Single Sign-On, and setup the means to deploy the app to production. I have also recently set up our project to use Node and Webpack to allow for Typescript development and testing, along with starting to work with Vue.js. I also frequently practice pair programming with fellow developers and am even responsible for Interviewing Drexel Coop Candidates.

Imalogix receives exams for CT scanners, and the technician fills in 3 fields (triplet) to provide details about the intent of that scan. This unique combination of phrases needs to somehow be mapped to a standard protocol (e.g. Head Neck with IV Contrast), so that we can group thousands of exams together under a limited set of categories that the user can set thresholds to alert when those exams exceed some amount of radiation. I designed and wrote a program that parses these phrases for keywords, and working with a Machine Learning developer, provide recommendations for an Imalogix employee to more easily map these triplets to protocols. This is accomplished using a programmatic thesaurus of synonyms mapped to keywords, and several other creative solutions that are still being improved today. The end goal is to have most of these triplets be automatically mapped by AI, and as of writing this, approximately 10-20% of them already are.