alt Hacker NewsDevOps is dead because it's run by a bunch of ops people who don't know how to do dev and a bunch of dev people who don't know how to do ops. The only tooling problem is that a bunch of companies created "DevOps tools" that then get dictated to use: K8s, terraform, etc. The only way this works is if you build the application to fit within those frameworks. Writing an indexer that is massively parallel and is mainly constrained by CPU/Memory. Instead, you have devs building something that gets thrown over the fence to a devops team that then containerizes it and throw it on K8s. What happens if the application requires lots of IOPS or network bandwidth? K8s doesn't schedule applications that way. "Oh you can customize the scheduler to take that into account". 2 years later, it's still not "customized" because they are ops people who don't know how to code. If you do customize it, the API is going to change in a few months which will break when you upgrade.
Replies
There are not very many ops people who cannot code. Especially these days. I spent at least the last 20 years doing ops. Ops people are HIGHLY motivated to create things that DON’T FAIL. However, ops teams are often blocked by MANAGERS from doing essentially development in the prod environment. I’m talking about tools and scripts. At the places I’ve worked with the highest uptime, it was because ops had an unlimited, unfettered free hand.
Remove the handcuffs from your ops team and your reliability will SOAR.
I would say, in pre-CC (pre-claude-code), this might seem like a daunting task for average DevOps engineers. But in post-CC, there is just no excuse to fret from such challenges.
EDIT: lol - I am getting downvoted for suggesting some DevOps engineers will actually be ready to take on tasks that were previously more intimidating. I really hope those folks are from the never-coding-agent camp. When I refer to. reliance on CC or Codex, I meant being engaged at a wholesome level with AI -- not blindly one-shotting solutions. This means having the patience to understand the complexity of the system, the criticality of its downtime in the overall architecture (in this case it's the k8s controller), ability to learn the codebase, using the right MCPs to delve into all the details needed for testing changes locally etc). These are system-level skills and barely overlaps with just coding skills.
Would you say it's truly dead or that it fails to meet the performance bar you've described?
The reality is that most devs do not consider a holistic picture that includes the infrastructure they will be deploying to. In many cases, it's certainly a skill issue; good devs are hard to find. And to flip the coin, it's hard to find good ops people too.
The reason DevOps continues to linger, however vague a discipline it is, is because it allows the business to differentiate between revenue generating roles and cost center roles. You want your dev resources to prioritize feature work, at the beckon of PMs or upper management, and let your "DevOps" resources to be responsible for actually getting the product deployed.
In essence, it's a ploy to further commoditize engineering roles, because finding unicorns that understand the picture top-to-bottom is difficult (finding /top/ talent is difficult!). In this way, DevOps is well and alive, as a Romero zombie.