Co-Founding Kubernetes with Microsoft CVP Brendan Burns

**Dan Lines** (0:00)
Conor, what do you want to do here? I couldn't get off the Kubernetes story.

**SPEAKER_2** (0:03)
No, this is great, I'm not asking to. Brendan, is there any way we can get 30 more minutes on our calendar at some point next couple weeks to finish this up, maybe?

**Dan Lines** (0:11)
Brendan Burns was so interesting, we had to have him back again this week. We'll talk about what he's up to at Microsoft, and he'll share the founding story of Kubernetes and what he's learned from it.

**SPEAKER_2** (0:22)
This episode is sponsored by LinearB. Accelerate your development pipeline with data-driven engineering metrics, continuous improvement automation, and project visibility while cutting your software development cycle time in half. Sign up for your free demo at linearb.io. Mention the Dev Interrupted Podcast discount for a one-month free when you sign up for an annual Pro membership.
Have you heard about Interact yet?
Our first Dev Interrupted conference is built by engineering leaders for engineering leaders. It takes place on September 30th. Learn more at devinterrupted.com/interact.

**Dan Lines** (0:56)
Hey everyone, welcome to Dev Interrupted. I'm your host, Dan Lines. And today I'm joined by Brendan Burns, Microsoft corporate vice president and co-founder of Kubernetes. Brendan, thanks so much for joining us.
Pumped to have you here. So prior to your time at Microsoft, you spent eight years at Google and were the lead engineer on Kubernetes and co-founder of Kubernetes really. Freaking awesome, man. How did the formation of Kubernetes come about?

**Brendan Burns** (1:26)
Yeah, it's interesting story. So I was in web search for four years. So I worked on Google web search, special focus on low latency indexing news and blogs and social media and like anything where you needed to get it from published into the index really really quick. And not just like wait and crawl, crawl once a day or crawl every 12 hours or whatever.
You post something on Twitter, you expect it to be searchable immediately. Subsecond.

**Dan Lines** (1:50)
Yeah, yeah, that's crazy.

**Brendan Burns** (1:51)
It's interesting.

**Dan Lines** (1:52)
Yeah.

**Brendan Burns** (1:52)
It's a fun problem. It was a really fun problem, but there was a big rework and it left Seattle where I worked at Google Seattle. I mean, I grew up in Seattle. So I switched and at the time we looked in cloud was really, they had a lot of white space. So it was really attractive. As an engineer, I find myself gravitating towards the places where there's lots of opportunity.

**Dan Lines** (2:11)
Sure. Yeah, it makes sense.

**Brendan Burns** (2:12)
Conversation. And so it had a ton of white space. And so we gravitated into cloud as a team. I was leading a small team. I was leading a team of about seven engineers at the time. And we came over and through doing cloud, I came to have a sense for the way that people are building things out in the real world as opposed to the Google world. Horrible, to be honest, right?

**Dan Lines** (2:32)
So you would hear like from a friend or something, like here, hey, how did you construct or deploy or release or this project? And what you heard back was, whoa, that's like behind the times or something? Or how did that?

**Brendan Burns** (2:43)
Well, I mean, it was people, people were using Chef and Puppet.

**Dan Lines** (2:45)
Yeah, I remember.

**Brendan Burns** (2:46)
A bunch of SSH and some bash scripts. And it wasn't so much that it was like they were suffering, like broken glass suffering. It was that it just didn't seem to work very well. Like 85% of the time, it would work. And I took to fail and it was heavy and slow and all this kind of stuff. And at the same time, Joe, Beta and Craig and I were all working together and Joe and Craig had created the Compute Engine, which is the virtual machine product for Google Cloud.

**Dan Lines** (3:10)
Yeah, I remember Compute Engine and then, okay. So they were looking for something you're saying?

**Brendan Burns** (3:14)
You get to a point and like you've built it and you've shown it and the other people are gonna take it over and they're looking and they're thinking about what comes next.
And I've been working a lot in the config management space, like CloudFormation-ish stuff or Terraform-ish stuff. And then actually looking at Chef and Puppet Insults and all this tech a lot too at the same time. And we all collectively found Docker right in the early days of when Docker happened. And we looked at Docker and it was like, wow, this is really, this is such a necessary prerequisite because in order to get people started, they have to be motivated to build container images and run containers. And so here's this amazing startup that has come along and built a case for why you should care. In all of our plans, step one is convince people to care about containers. And all of a sudden it was done, right?