AI

Where the Edge AI course told you why compression matters, this course opens the hood. You'll work through the real mechanics of quantization, pruning, and knowledge distillation: the scale-and-zero-point math that turns floats into integers, the difference between pruning that needs special hardware and pruning that doesn't, the temperature parameter that controls how much a student model learns from a teacher's uncertainty, and the order these techniques get chained in a production pipeline. The course ends with the question every compression project eventually has to answer: how do you know the smaller model is actually good enough to ship.