ML and AI are some of the most commonly used workloads in Python. We want to make sure the major libraries are running efficiently on RISC-V to make it a competitive platform for these workloads.

Numpy is the fundamental package for scientific computing with Python. It relies on native BLAS implementations for some of its compute, but some other parts use Numpy-specific implementations. We want to make sure these Numpy-specific components are fast on RISC-V.