ml systems explained by someone who came from backend engineering, not ml

i've spent the last few years building distributed systems on azure, using services like key vault, blob storage, service bus, fabric, data factory, functions. outside of work i've been reading up on