This is kind of funny, the idea that there’s no jobs for engineers in infrastructure… I sometimes hear about NoOps. That’s just neglecting to mention the team of engineers behind Google or any other cloud. There’s a gob of ops engineers running all the ‘serverless’ clouds. And at some point, a company gets big enough that they just do a lot of private cloud — Google, Facebook, Amazon, Microsoft all have considerable talent on their infrastructure teams. That’s the place where infrastructure engineers have jobs, and DevOps is fantastically profitable, if invisible to the cloud users.
Serverless is practical for companies of a certain size, especially if they don’t have intense compliance issues. Yet it’s important to point out that the services offered by public cloud are profitable for a reason. It’s because they own the infrastructure engineering talent to host the platform. The cost for everyone else is higher, because Ops engineers can go work in house where everything is on their own terms; they no longer need to cater to specific Devs. If running a private cloud didn’t have advantages, Facebook wouldn’t do it. Yet they are heavily invested in their own datacenters, and even have the Open Compute project for publishing server hardware.
It’s kind of cool that you don’t look deeper into the cloud where you deploy. Serverless is anything but — there’s a lot of servers back there. You’re showing off golden handcuffs as though they were your favorite bling. Maybe that’s the frontend life.
