I'm not the author, the guy you're replying to is. The article that you quote says that crossing function barriers and joint compilation are issues with numba a

I'm not the author, the guy you're replying to is. The article that you quote says that crossing function barriers and joint compilation are issues with numba and cython. So does the comment you were replying to. You've failed to articulate where the perceived discrepancy is.