basically the idea is you do not need to configure the Experiment manually, it is created when you actually develop the code / run/debug it, or you have the CLI taking everything from your machine and populating it