DevOps Isn’t Dead: re:Invent Shows Cloud Native and Platform Engineering Rising | Shimmy Says Ep. 36

Hey everyone, it's Shimmy, and welcome to another Shimmy. Says, it's a little different. Shimmy says, I'm in Vegas.

You know, you know what they say? What happens here stays here. Well, not always with Shimmy.

I want to tell you about what I've seen here and what we've, uh, been talking about here. So I'm here at AWS reinvent, as you might imagine. And you know, I wrote a couple of articles on, I guess it was Monday when I first got here and, and saw the, uh, agendas and the keynote previews and so forth, and it was clearly obvious AWS was going all in on agent AI and for good reason, right?

They've been, they've been getting dinged for maybe being a little bit behind some of the other hyperscalers in terms of their a ai, uh, strategy roadmap offerings, right? Google has an amazing vertically integrated stack with Gemini and, and all of the hardware and, and just everything that Google, you know, the assets. Google has Microsoft with GitHub owning developers.

Uh, their AI in the open AI investment seemed to have a very well developed and, and already rubber meaning the road type of plan for ai. Oh, now, A AWS has been talking about ai. They had Bedrock and q and they've of course been in a strategic investor in Anthropic, but I think the perception was they were laggard a little bit and, and I think they wanted to hit that head on.

And so when you got here, it was clearly, you know, agent AI all the time, all agent AI all the time. Now, what a difference a day makes, or in this case, two or three days over the last two or three days, I've had a chance to do a whole bunch of videos. If you haven't caught 'em on text, drunk tv, go check 'em out.

We stream some of them, some of 'em are recorded, but I've also had the chance to speak to a lot of people, see the floor, see what's going on, and I'm here to happily report that DevOps cloudnative platform engineering are alive and well here at AWS reinvent. Um, look, as I said, ai, there's a lot of noise, and it is the lead story on, on just about every article or video you might be seeing from here. But when you peel that back and you go a little deeper, there's a lot of news, a lot of information, a lot of sessions around DevOps, cloud native and ai.

Let me, let me hit some of this stuff for you on DevOps, you know, know they, AWS announced three key agentic ais that they were coming out with. One was DevOps, uh, an AI DevOps agent. Now, is it the be all end all for DevOps?

No, but it's a start. It's part of the kiro developer, uh, platform that they're, they're, they're moving forward with. And I, you know, it, it has a lot to do with testing code or moving code along in your CICD.

It's great, right? It's a fine start. I, and, and from what I've heard in briefings and so forth, it's really just the first of several others that we're gonna see that are really here.

Not to replace the DevOps engineer, but to make the DevOps engineer's life easier to make his role actually more important. So, net net, I think it's a good thing for DevOps. On top of that, there was just a lot more talk about DevOps in general, right?

I just, about every DevOps vendor that I know of is out there on the show floor and they've got nice crowds around their booth. Um, you know, there's no one walking around here with DevOps is dead nonsense, and it's, it, it was heartening. I I think DevOps has made a real comeback here at AWS reinvent this year.

And I, I think that's a great thing. Let me move over to cloud native ai or no, AI cloud native remains the preferred platform for organizations looking to modernize and transform their infrastructure. If you're moving from a data center to a cloud, and you don't have to be all in, right?

That's a lesson here. No one's throwing out their old data center or moth bowling it or turning it into a video studio. They're keeping stuff in the data center, but they're moving some payloads to the cloud, to the hyperscaler.

What does that mean when they move it to the cloud? Generally, they're taking the opportunity to modernize and transform. They're moving to a, a, um, Kubernetes container based a architecture.

They're moving to a microservices based architecture. And, and that's important. They're moving to a microservices based architecture.

They are using observability, they're cloud. The cloud native stack is the compute stack of choice. If you are modernizing your stack, and I'm hearing it not just from AWS but even, you know, our friends at Seus who, who sponsored, uh, our video program this year, they're all in on this too.

They, because they realize how important cloud native is, well, they, they have a rancher, right? And with rancher, you can now you can manage multiple Kubernetes clusters, which for those of you who are in cloud native engineers watching it, let me just tell you, it's hard, it's hard to manage Kubernetes. It's even harder to manage multiple Kubernetes clusters.

So the, the, the Souse Rancher, uh, announcements here were key and, and multiple Linux. You know, AWS is about choice. Uh, and, and I think that came through loud and clear, giving you lots of choices in your cloud native journey of how you want to do this.

'cause there is more than one way to skin the cloud native cat, so to speak. Um, and what's nice about it is cloud native GI ops, dev, DevOps, all are, you know, they're all integrated playing nice together. And, and I, I think, and they all play nice with ai, right?

This is a case where AI's not going to make, render these obsolete. AI actually enhances the capabilities of your DevOps team, of your cloud native architecture. And let me throw one other stick on the fire too.

Platform engineering, you know, for the last couple years, platform and engineering has been kind of the new kid on the block. Some people say it's a marketing, we're saying it's a marketing gimmick. Some people were saying it's a fad.

You know, some people were saying it's a niche kind of thing, and it's just really another word for first, they were saying it was another word for, for managing Kubernetes the last two years. They were saying, well, it just means you're managing an IDP internal developer platform. But at the, at the end of the day, make no mistake, and the age of AI platforms where it's at, you wanna do a a point solution.

You AI can spin up point solutions all day long till the cows come home. But you need a platform for these things to run on. And whatever your platform is, whether it's backstage or an AWS platform or anyone else, platform engineering is the way you manage it.

And they're well represented here. It, it's taken on a bit of a maturity. org.

We're here doing some supplemental sessions and so forth. Um, I, I think a message back home to the platform engineers is, platform engineers is part of this. It's integrated into DevOps and Cloud Native.

And the three of them work real well together, and the three of them work real well with AI as well. So anybody who, who, um, you know, might have been thinking that DevOps, cloud native platform engineering are taking a back seat as we all transition to this AI future would be mistaken because AI doesn't replace these things. It grows on top of these things.

And there's a lesson there for all of us, right? Lot of, lot of, uh, announcements out here around hardware. All of a sudden, hardware, you know, I've said it before, hardware's sexy.

It hasn't been sexy for a long time, but today it is. If you could talk the latest Graviton and trains and Blackwells and all these names, good for you. But there's that hardware layer, there's the software layer, there's what we've learned in it and the processes and, and, and mod modality modalities that we've built up for decades.

And AI enhances those, speeds them up, makes them better, but doesn't replace them. And there's a lesson there for all of us. It doesn't replace you either, right?

AI is not necessarily taking your job. Don't believe that hype either. It enhances you if you let it, and I'm gonna leave you with that here from Vegas.

'cause anything more, I'd be, you know, I can't, what happens in Vegas, I can't tell you too much about. But for now, until next week, this is shimmy for Shimmy says.