Flyte enables production-grade orchestration for machine learning workflows and data processing created to accelerate local workflows to production.

Flyte

Hello!
I have a question regarding running workflows localy vs in a cluster:
Are workflows designed to be agnostic of their execution context? How do you handle local vs cluster data input?

Yes they can be completely portable unless you directly depend on some technology 

The context is auto determined by the cluster in which they are run or by the data paths for example &lt;s3://:|s3://:&gt; or / or gcs:// etc

If th code just uses Flyte objects like flytefile etc or dataframes - it should just work