i have been watching karparthy's neural networks zero to hero youtube playlist and this is where i will be capturing all my notes

ep 1 - building micrograd ↘

-> either backprop is easier than i imagined, or karparthy is a goated cs educator

-> i should implement micrograd myself, seems ^{(deceptively?)} simple


ep 2 - building makemore ↘

-> bigrams are very simple, i accidentally had made something much similar in undergrad without knowing they were a thing

-> karparthy talks in first person plural which feels kinda nice