My final role at Amazon was inside AWS, in the Migrations Service group. I worked on the AWS Discovery Agent, a C++ agent designed to help customers get a better understanding of their on-premises data-centers. Agents capture server configuration, system performance, running processes, and details of the network connections between systems. Agents provide the data both to the customers and to the AWS Migration Hub. This allows customers to track and migrate applications to AWS.
For a while at Amazon I worked on services for the World Wide Operations Human Resources Talent Management team. The most prominent service I worked on was the Talent Algorithm (TA) project. TA was built as a web interface around a workforce machine learning model. The algorithm could estimate needed headcount. Utilizing this project, teams were better able to predict hiring in the future.
My first role at Amazon was as a Software Engineer on the Core AI team. The service I primarily worked on was the Real Estate Evaluation Tool (REET). This tool was created to help the real-estate team responsible for placing Amazon fulfillment centers (FCs) across North America and Europe. REET placed an aggregated list of all Amazon real estate on an interactive map that allowed for simulations of new FCs. There was a economic model that underpinned the web-interface. This model took many labor related metrics from a variety of sources and created a score on the availability of labor for every zip code in the United States. This information was then given to the Real Estate team to use in purchasing decisions.