I was a 1st employee at a fairly successful startup and am now an "early stage founder". The experience from one doesn't carry over as much to the other as you'

I was a 1st employee at a fairly successful startup and am now an "early stage founder". The experience from one doesn't carry over as much to the other as you'd expect. 1) Somewhat true, it's definitely the case initially but you'll see politics develop if the company gets above 15-20 employees. 2) You'll half ass everything. It's nice to get exposure if you don't know what you're interested in yet but it will be hard to go deep in any subject. If you're doing ML the startup will probably have no data, process and infrastructure for you to be effective. At Google they have all the data you can dream of, 1000s of people to label things if you need them, unlimited hardware and tooling to run and manage experiments. 3) You'll learn a lot more about functional organizations at Google than as an early engineer at a startup. You'll also come out with a network and resume that will make raising money and hiring a lot easier.