That makes the sentence quoted by GP sound like a category error to me. Developing better algorithms and developing more useful models to run on those algorithm

That makes the sentence quoted by GP sound like a category error to me. Developing better algorithms and developing more useful models to run on those algorithms are not an "either/or" situation.