Java or Scala - as a second language for an ML engineer?

Question

Java

Which one to choose for the next 3 years for an ML engineer in addition to Python?
The main scope of the application is pipelines (asynchrony / parallelism), output of models to the production (microservices / containerization) ...
Both at once are not an option (there are only 24 hours in a day) :)

Answer 1

Spark is more convenient to use with Scala. And what is ML without Spark?

Answer 2

Scala -> for a pythonist there will be a lot in common. Plus, types are derived from operations, you won’t need to constantly describe them, as in Java.