Hi @stucas ,

Adding following configuration spark.databricks.delta.schema.autoMerge.enabled = true
to the DLT pipeline will allow new pivoted columns to be merged into the target table automatically. However, DLT still requires a defined schema at initialization, so this won’t fully solve the issue of handling a completely dynamic pivot schema. It can help if your pivoted values evolve gradually, but DLT doesn’t natively support unknown schemas upfront.