Great writeup, generalisation and data seperation is difficult but will more often than not improve code. Learning how to structure your code around general con

Great writeup, generalisation and data seperation is difficult but will more often than not improve code. Learning how to structure your code around general concepts has the advantage that there many low-level vectorized libraries available increasing compute efficiency. This is especially true for numerical programming (numpy), stats (scipy) and ML (torch, tf, jax), but the difficulty is often in finding the name of functions to corresponding concepts. Better get good at formulating queries against the API documentation too!