How can we store feature transforms in a feature store?

You can use SageMaker Data Wrangler to create data transformation workflows. Once your data is prepared, you can build fully automated ML workflows with Amazon SageMaker Pipelines to publish the transformed features in SageMaker Feature Store so you can share features across your team and others can reuse them for their own models and analysis. SageMaker Feature Store keeps track of the metadata of stored features (e.g. feature name or version number) so that you can query the features for the right attributes. Also if you like to monitor your features in production for any drift you could use data profiling integration with Feature Store, more details in this blog post.