This is horrible LLM slop, my god.

Winced my way through “Convolutions are in CNNs (it’s literally in the name, Convolutional Neural Network)”, then had to stop.

It’s honestly offensive to me. It doesn’t even make sense on its own terms. For some reason we fly from LLM inferencing to toy MSINT to convolutions with __0__ transition or sense of structure.

Aside from the verbose AI slop it's an interesting hobby project for exploring FPGAs. But it doesn't do anything you can't do on CPU by using a model that's small enough to fit in cache. In terms of practical use you'd be better off implementing a minuscule model using vector intrinsics in your favorite systems language.