I personally like scale as I work in ml and that helps in making datasets large. You can do a lot of ml on small datasets but life is much easier when you get t

I personally like scale as I work in ml and that helps in making datasets large. You can do a lot of ml on small datasets but life is much easier when you get to work with what feels like infinite data. It was nice having models train on trillionish data points at a prior company. The opposite end I worked at a startup that had a couple thousand data points and dataset size was quite annoying especially in not having enough variety.