DevOps is the Philosophy, Platform is the Practice | Humanitec's Kaspar von Grünberg

**Kaspar von Grünberg** (0:00)
Our industry is the only industry that is not actually driving towards specialization.
Since medieval ages to now, every industry has always specialized. We're the only f***ing industry in the world that is actually working against specialization. And where I really have a problem with these fundamentalist approaches or viewpoints is that many of them just have never really worked at scale.

**SPEAKER_2** (0:30)
Attention, engineering leaders. Join the Dev Interrupted community for an exciting podcast streaming on April 4th and 5th. Listen as Nick Cobb, head of product and engineering at Kite, discusses the Silicon Valley bank crash, how Uber's engineering team lost its way, and the challenge of AI self-driving. You'll also get exclusive insights into Kite's experimental product innovation playbook.
Record in front of a live audience at the Lead Dev Conference in New York City and hosted by me, Conor Bronsdon.
This in-depth podcast is not to be missed. Mark your calendars and join the conversation. Register today with the link in the show notes.

**Dan Lines** (1:03)
Hey, what's up everyone? Welcome to Dev Interrupted. This is your host, Dan Lines, co-founder at Linear B. And today we're joined by Kaspar von Grünberg, founder and CEO at Humanitec. Kaspar, welcome to the show.

**Kaspar von Grünberg** (1:20)
Dan, thank you for having me.

**Dan Lines** (1:22)
Yeah, awesome to have you here today.
Before we got on, like kind of prepping for the episode here, I checked out your LinkedIn profile. I'm like, okay, who is this Kaspar? Under your name, you have platform engineering with a heart as they do now, right?

**Kaspar von Grünberg** (1:38)
Cheesy, right?

**Dan Lines** (1:40)
No, I liked it.
And so I knew, okay, this is gonna be like a pretty cool conversation. You're the CEO, you're the founder at Humanitec. We're gonna get into all of that. You're doing some really, I think, cool things in the platform engineering space. For me at Linear B, we also have a lot of platform eng, DevOps people using our platform, so we kind of have some synergies there. What I wanted to start out with is, I think for a lot of people, platform engineering, we heard of DevOps forever now.
Platform engineering, can't really tell, is it a buzzword or it's an actual career? Let's start there of how would you define platform engineering?

**Kaspar von Grünberg** (2:28)
So platform engineering for me is the... So the platform engineering team is the team that looks at all of the tech and tools that your company has and then treats them as a product and says, hmm, if we think of our user, the user is the developer, but how do we need to structure those tools? How do we need to organize everything to bind them into delightful golden paths that they can use and that are reliable, that they can actually consume with low cognitive load and they can reach what they want to reach, right? Write great code, build great business logic without actually having to wait for operations tickets or understand complex, I don't know, terraform code.
That's what we need platform engineering for.

**Dan Lines** (3:17)
And why do you think it's kind of heating up right now? You see people putting it in their titles. I'm a platform engineer now. You see people writing about it.
Why do you think it's a hot topic?

**Kaspar von Grünberg** (3:30)
I think two answers. Answer number one, there is an underlying trend, if you want. If you look at the uptake of cloud native, if you want, right? You had the 2006, Verner Vogels says in a blog post, you build it, you run it.
Amazon starts, all of these things.
But if you look at when the majority of teams actually get mainstream cloud native, that's like 2019, 2020 Think about when was EKS available in the rest of the world? I think that was like 2019 or early 2019
So if we look at the large teams that have been cloud native for a while, Google, GitHub, internal, Airbnb, after a couple of years, they all realize that shit is complex. That idea of, let's just throw everything at the developer, that doesn't fly, like the FTP request, well, at times they're over. It's really complex. We got to build a platform.
And there is this moment 2017 where Jason Warner, the SVP engineering at Heroku, moves over to GitHub and looks at the situation and says, this is really inefficient. It doesn't make sense and starts building a Heroku, but on top of, I think, Amazon. And to a certain degree, that's the start of what we now call it an internal developer platform. And that's reason number one. Reason number two, a lot of us have spent a lot of time starting to articulate that. So when we started the term internal developer platform that didn't exist, platform engineering barely, a lot of these terms are very new. And we said, OK, well, let's actually get that ball rolling. And we spent a lot of time working with different folks to articulate that and say, like, hey, this is actually, this is more than a buzzword, it's more than a title. And we built these large communities and they're really large now, like PlatformCon this year, I think we just got the numbers. We were expecting like 150, 200 proposals. I think we, in the end, got 500 or something. And we're expecting over 15,000 people. And this is really like a rolling ball.