Today, you cannot directly create batch dataflows to apply to data in memory. The nuanced answer is that you would need to Export an in-memory table from your Operationalizing cluster and import into your Designing cluster in order to create a batch dataflow using the data. This could be a case where I might suggest you parameterize your dataflow for easy reuse.
What are you trying to accomplish? I am glad to submit a feature request on your behalf for the ability to add in-memory data and aggregate variables to tables. Detailing your objective would help our Engineering team vet the feature request for addition. Let me know what you are trying to do, and we’ll put our heads together and figure it out!