i started with jvm implementations and then added ...

# datascience

[kyonifer]

i started with jvm implementations and then added a pure kotlin one later that works for all the platforms (because doing linalg is easier than reinventing spark from scratch!)... but barring that platform-specific backends would also work