as part of our data science use cases we sometimes have the requirement to do live inference / deploy a model as API endpoint to be called from a dashboard, for example.
