Data Scientist
My duties ranged from data collection to model hosting, I was responsible for every step in the pipeline for my given projects. I had a total of two projects during my time with Magnus, one was an estimated time of arrival (ETA) model and the other was a truck driver churn model. Given that the data science team was new, my co-worker and I established a data lake, model endpoints and functional APIs for our projects, all which were constructed to be scale-able. The company was shifting to half-integrate Amazon Web Services (AWS), thus our pipeline was constructed using many tools from AWS such as Glue Data Catalog, Athena, Sagemaker, Single and Multi Model endpoints and Model Monitor.