Disclosure: I work on Google Cloud. Nearly all VCs and Accelerators are on our list of partners at cloud.google.com/startups where that up to $100k in credits

Disclosure: I work on Google Cloud. Nearly all VCs and Accelerators are on our list of partners at cloud.google.com/startups where that up to $100k in credits offer lives. You would not believe how contentious it was in the early days of GCP to fund this... Unicorns take a while to grow, but I still think Snap was one of the most fascinating stories, as they were entirely on App Engine and Datastore well into their unicorn days. Is Niantic interesting? Shopify? Presumably Spotify stopped being a unicorn startup and is now just a successful company. To your other point, lots of ex-Googlers choose GCP, which I mostly put down to "I don't worry the way others do". If you've been at Google, you already know and trust a lot of the underlying systems and people. Adopting BigTable, Spanner or Dataflow doesn't sound scary to you, or make you worried that you won't be able to hire people that "know GCP" the way that "everyone knows AWS". Similarly, you "know" that Google isn't going to shut it down. Others don't necessarily feel that way, and frankly we just don't even get considered. I personally think that's an interesting signal for a startup: if they do a technical evaluation and choose GCP, rather than sticking with AWS as the default. The ex-Googlers choosing it because they still have friends at Google isn't a particularly great business plan for us, but they do provide great feedback! (Particularly people who say "Umm, I used to work on < X >, can't you just enable that for the public version? I've got money").