• Partnered with Ernst & Young for a final year university unit to develop a web-based solution that utilized machine learning with computer vision within an Agile SDLC. Main technologies used were Docker, TensorFlow and Flask.• I served as the lead developer for the web-based application. I oversaw most software related aspects of the project in conjunction with other responsibilities. My main responsibilities were the designing, building and configuration of the application to meet business requirements and application requirements. I was also directing the development team in the design, development, coding, testing and debugging of the web-based application for the relevant sections that I had the most knowledge in.• I worked closely with the Ernst & Young staff to understand business requirements of the software and helped translate these into technical requirements for the team. Apart from this, I was also heavily involved in the documentation process, which required detailed commentary to both the university and Ernst & Young staff on many aspects relating to the project and various sprints.• The software related aspects of the project that I worked upon include the construction of a web-based application and all relevant variables, code evaluation, debugging, data pipeline construction, evaluation of deep learning models and integrating the machine learning into the web-based application.• The software was created to analyze computer vision through a variety of media sources with the ultimate goal of image recognition. It utilized a state of the art pre-trained convolutional-neural network model which I codeveloped. The model that was 99.9% accurate on the testing data with a 10-fold cross-validation on 3000+ images. Upon deployment in a pseudo-realistic setting, the model performed as expected, with the Ernest & Young staff being impressed at the quality.