An image detection model or an audio detection model is only as good as the labeled data set used to train it. In order to generate more refined models & more specific models for our clients, we decided to create the 'Silk Deep Learning Tool' that allowed everyone in the company to rapidly label large amounts of data.
Prior to creating the tool, only our engineers were able to label data using our CLI. After the tool was created, everyone in the company was able to participate. It was an exciting project to work on and really gave me a deep dive into the nuts and bolts of machine learning and how detection models are created.