A lot of people use databricks for the analytical and ML workloads (leveraging it as a sort of extended spark environment)