I’m an ex-GCP engineer, and have also been a customer on the platinum support tier of GCP, overseeing 10k+ Cloud resources. I just quit my SWE job to address t

I’m an ex-GCP engineer, and have also been a customer on the platinum support tier of GCP, overseeing 10k+ Cloud resources. I just quit my SWE job to address this very problem. Obviously, our team holds some conviction about the vendor risk and need to derisk. Here’s our shameless plug and our two cents on the space. We’ve built a cloud agnostic search engine as well as means of provisioning in a cloud agnostic way. The premise is that, ultimately, a VM should just be a VM. Cloud vendors could be viewed as utility companies or commodities, with fungible offerings. Currently, there are still vendor specific details you have to be concerned with, so lifting and shifting from one cloud to another is not as straightforward as we’d all hope. We think every CTO should take a cold hard look at IaaS vendor risk. Hashicorp may be better suited if you have the engineering capacity for infrastructure-as-code. We’re more GUI and low-code centric. Spinnaker is also a great cloud agnostic tool if you’re mostly just dealing in VMs. There is some time however. Large cloud customers have multi-year contracts where they pre-pay for resources. It would be legally very costly to negate those contracts, as opposed to consumer products such as Stadia. Waving my hand a bit here, but if there was a way of speculating the trend of outstanding contracts, it would be a call to action to derisk sooner.