hi @szymon_dybczak - thanks for responding. The error is saying its in the column name. I think it has to do with the space that's in the 'child' column. However, I've been asked to not flatten the pyspark dataframe and just ingest the 'raw' data, ie. leave the nested columns nested and not flatten. I feel like that might be the issue because if I convert the pyspark to a pandas dataframe I'm able to load this without issues to my table. Is there a way to not 'flatten' and just load to pyspark dataframe?