It's been great to see Microsoft embracing the R language on Azure, being able to easily operationalize R assets is changing the way organisations think about their analytical workloads.
While it is trivial to publish an R model as a web service in Azure Machine Learning, there is still no easy way to integrate this within standard ALM processes and tooling. For one of our customers this was a big issue. They use Azure DevOps and wanted a solution that would allow them to version their R models and deploy them across dev, test and production environments.
We solved this problem by creating a set of reusable scripts, designed to be plugged in to ADO or any similar automated deployment tooling, that can deploy R models to Azure ML.