Dataiku said this support lets data developers/scientists write Spark-native code-based ML transformations while producing data applications, adding to the existing support for Python, R, SQL, Hive, ...