Hi @Chris Nawara​,

I had the same issue you had. I was trying to avoid the apply_changes but we in the end I implemented it and I'm happier that I expected hehe

and if you have any additional standardization columns that you need to implement, you can simply read from the apply_changes table and generate the final table.

My logic is like that

readStream -> dlt.view based on the dataframe -> dlt.create_streaming_live_table -> dlt.apply_changes (stored_as_scd_type=2) -> dlt.table (I had to create an additional table because I have few columns to calculatated based on the __START_AT and __END_AT provided by the apply_changes)