Reducing Toil and Error: Pavlo Baron’s Solution for Software Delivery

Hey everyone. Welcome back here to Text Drunk tv. I'm really happy to have this next guest on.

It's a new, a new company we haven't featured before. It's his first time here. So let me introduce you all to Pavlo Baron.

Pavlo is the co-founder and CEO of a company called Platform Engineering Labs. Pavlo, it's great to have you on. Nice to meet you.

Welcome. Thank you. Thanks.

Thank you so much for having me. It's our pleasure. Pavlo, before we jump into platform engineering labs and something called Fori fori, I, I want to talk a little bit about you and give our audience a sense, as I mentioned, you're the co-founder and CEO of the company.

Tell us a little bit about you. Alright, I'm looking back at a, uh, very long career in, um, well, broader IT space, let's call it like that. I'm, uh, pretty much a veteran and, um, well, I basically was in every single role you can imagine in different kinds of companies, different domains, always staying curious, going after the means.

I didn't know so far was architecting, creating platforms, leading engineering teams, uh, building teams, building startups. So the most recent, uh, achievement is, uh, co-founding and, uh, a co-founding, a company called Instana, uh, that exited to IBM 2020. And that stayed with IBM for a few years, um, to, to have an internal role as a, um, in the, in innovation space.

And, um, yeah, I represent the company. So I need to tell a few words about my co-founder. My co-founder, Zach Schneider.

He is a good friend of mine. And, uh, we worked together at Inana already and we always, always kept talking about, um, well, how to change the things in the space we're going into now as a company. So Zach lives in, uh, sunny Traverse City, Michigan.

And, uh, his, uh, most recent activity was, uh, he was one of the core engineers of iCloud at Apple. Really? Right.

Very cool. Very cool. And he was at Stanner as well?

Yes, sir. So of course, well, I'm personally familiar with Instana. We worked a ton with Instana, you know, prior to the IBM acquisition.

com, I think. But, um, interesting. So were you a co-founder at instana?

Is that the story? Yes, I'm, I'm co-founder and CTO and I invented big chunks of it. Very cool.

And Zach also was a co-founder there. He just worked in Inten. Yeah, He was my first hire actually in America.

Really? Yeah, because, uh, it was 24 7 with humongous ingress, a lot of data flying by. And, uh, in the early days I took over also the responsibility to actually run this thing, not only to build this thing with the team, and then at some point I had to make a decision to actually cover the US hours because we had us customers and Sure.

I was myself in Europe like I am right now. And, uh, this is how I met Zach through another friend and, uh, it turned, uh, into a really good relationship and friendship. That's fantastic.

What a great story. You know, my, my very first company I co-founded it is back in 95, 96. Uh, I didn't know what it was at the time, but it became known as web hosting.

I thought I was a virtual landlord. I was just buying hard drives to store more websites. Right.

And, and we, it, I had the same thing in reverse. All of a sudden we had a lot of European customers and that meant we had to do tech support for us. It was the middle of the night.

Right. So, so after a few, a few months of staying up all night, you know, running, we, we ran Solaris, that's how long ago this was. Oh, yeah.

Right. Being trying to be a Solaris admin and then go to work in the morning and run a business. I said, this is crazy.

I need, and I did the reverse thing. We actually picked up a partner in Germany. Right.

Uh, who, who was a, he was a great customer. We gave him, basically gave him hosting for free and made him, made him our, you know, our tech support overnight until we went to 24 7. Right.

So familiar with that. Familiar with that. Very much so.

So Pablo, let's talk about platform engineering labs. No one, you, you've started enough companies, you know this, no one undertakes starting a company lightly. It's a commitment, right?

Right. It's, you put your life and your soul into it. Yes.

Tell us, what was the passion that drove, what's the mission of platform engineering labs? Well, the broader vision, we actually want to eliminate the toil, the unnecessary labor, the human error from, let's start with the infrastructure space, but generally everything that is about delivering software, um, the problem we are addressing is old and it's not solved. So Zach and I, as I said, going through very complicated, um, environment and, uh, you know, massive data ingress and a lot of responsibility for 24 7 platform running for a lot of customers on premise as well in SaaS, we actually learned a lot of things that we want to improve.

And, uh, we met a lot of people because I mean, you can imagine that when you are an a PM solution that gets, uh, a lot of customers and a lot of attention, you basically meet everybody in the ops space, right? And let's call it the broader ops space because there is just a little difference between the disciplines, but it's the same people with the, the same skillset. It is about people who are running software and, uh, we met a lot of them.

And, you know, there is so much pain out there with just a few things that are not properly addressed in this space that we just thought that we need to change that at some point, once we learned that, uh, uh, there's a, a lot of movement in the markets. For example, back then when HashiCorp acquisitions for IBM was announced, we just, Hey, it's about time to start it now because we know that, uh, this space is something we want to change, even if this is the last thing I do in my life. So that is, let's Hope not, let's hope not.

No, You and me both my friend, but this is the commitment. I mean, we definitely, we are, are on a mission to give platform engineers, and I'm talking about platform engineers as a species. This is finally a discipline that allows us to work the way best teams actually work, which is a few real good people and the few people who are just willing to go full stack as deep as possible, and who willing to go on call all night.

This is a special species, as I said, uh, this people work for a hundred other people. So basically one team is serving a hundred other teams. And, uh, that's the cool thing about the platform engineering because after all this cultural dancing around DevOps and whatsoever I am, I'm just seeing this as a chance to really set it up the way it was supposed to be.

But, and here's where a company comes into play. These people have to deal with tools that are completely outdated and, and, and have never been designed for this, uh, mode of operation. We're talking about develop experience or user experience from the, uh, from the late sixties.

I mean, everything is low level. Everything is completely detailed. You are the guy that is part of the tool, not a single tool they have to work with can allow you to lean back and, and, and, and let it go.

Um, while everything in the dev space is being constantly improved and simplified and abstracted, none of this is happening in the ops space. And that is exactly what we want to change. And, uh, we will definitely go one by one with the current focus on the infrastructure management because with no matter what tool we tried in the best, and you can imagine, Zach and I, we really went through all of them and we had a pretty big responsibility using them.

We tired and we, when we are tired, we talk about everybody else was tired. I know enough people in the space who are really good, but they're close to checking it all in and just going gardening. That is how bad it is.

And I wanna change that. That's, that's my mission. Good for you, man.

I love it. com, I, I understand. I feel your pain.

I understand the mission. I, I, I get it. Um, I also look at platform engineering, not as a replacement for DevOps.

You, you're still gonna do DevOps once you have your platform laid out, but it's sort of laying the groundwork that we can build on top of. I, I think one of the things we learned and, and you know, Pavlo, you've been around, I've been around, right? We, we, we see what, how this works.

One of the things we, we've learned in the DevOps movement is know onesies and twosies and small teams. It's, you can really control things and, you know, fine tune and dial in when you go to scale, Right? Scale's a different animal, a different beast.

And, and what works in a small team well, doesn't necessarily work at scale. Right? Well, and and to me, that's really the mission.

Build a platform that scales, right? That allows your developers, your DevOps engineers, your SREs, your, your qa, your security people to go as fast as they can go. Exactly.

Sir. Yes. I love that.

This is exactly the point. We have more kind of professions coming up in this industry. I'm talking about all this AI engineers and everything they get.

Sure. All of the infrastructure management built in, all of the security tools actually manage the infrastructure themselves. You are in the situation that you cannot manage everything with one tool.

And I'm hearing people using within the same setting, two to three different infrastructure management tools, that is absolutely impossible to, um, well, to call it being under control because everything is happening in parallel. And if you slow them down, actually that, that's, that's bad, right? I mean, you, you want, yeah, you want everybody, as you said, to move at at the speed they want and have to move it.

And what is the job of the platform team or operations team or SRE team, however we call them, of course it is to support them. But since this is exploding and it's gonna explode more and more and more, the more AI is gonna be used because everything is landing on your desk. Your only job is to get rid of routine work and put yourself in the driver's seat and always have a feeling of control.

And this is, this is exactly what our goal is with this first tool we are building regarding the infrastructure management. We want to put you exactly in dust in that driver's seat. Let everybody go at the speed they want.

But you are the one who control and manage things. I love it. You know, and, and you mentioned ai.

Look, AI doesn't do well in, in chaos. It'll do best in a, in a very defined environment, right. The more you can define it, the, you know, if it knows what the rules are, it operates within those usually.

Um, so I, I think it, it, it behooves us as well, if we're gonna go with an ai, everyone wants to go ai, ai, ai. I get it. If you have, again, you have a platform with guardrails, with clearly defined, you know, routes, then AI can go faster than any humans are, are gonna be able to go as well.

Um, so it, it all makes sense. I gotta do a little business housekeeping, pavlo platform engineering labs. What's the website?

It's platform engineering. We have a great, love It. Great.

Um, and company. It, it's, it's out of, it's, it's in it, it's out of stealth I should say. It's public.

It's, is it funded? I, I don't know. Yeah.

Have did you raise money? We have, we have investors. We're prese.

Okay. Uh, we have an announcement that is a few months old already of our pre-seed round. Uh, it's out there in the public, so it's not secret.

Mm-hmm. Very good. Um, all right, let's turn to, and I, you know, as we were talking off camera, Latin's, not my forte, they'll get a little pun there, Forte's Latin, but, uh, but for, is it forme or for my, it's forme.

Forme, Yeah. Lucky me. My daughter studies medicine, so, uh, she can teach me that.

Excellent. So for me is, is this new, uh, open source infrastructure is cloud platform that Platform Engineering labs has, has pioneered or has developed, and it, it's basically, you know, IT infrastructure is code built for the future tote. You know, that's at a high level.

Tell us more. Right? Infrastructure's code is a very interesting space.

There's a few players, it's not really crowded. This, uh, the space is dominated by, uh, Terraform. Terraform, uh, is developed by HashiCorp.

HashiCorp is a, is acquired by IBM. Uh, there's a few other players, uh, in the market with their own, uh, uh, advantages and disadvantages. But our goal is a little bit different than what everybody else is doing, doing.

We just looked at what is, what is slowing down those people who work for the other teams in regards to infrastructure management. There's a, there's a humongous divide between what those platform teams are doing and how developers are working. For example, I'm yet to see a developer, a regular developer who wants to go on call and wake up at 2:00 AM to fix their stuff.

They're not doing that. This is the responsibility of somebody else. We call them the operators.

We can put them in whatever discipline you want. Uh, it doesn't matter how we call it, even platform engineering. That is a role, it's a responsibility.

We want to support this role with a, with the adequate tool. And honestly, we learned, uh, for many years of experience, the best tool people work with and everybody agrees on is code. So what it, what is for me, for me is a hundred percent entire coding code out infrastructures code tool where tools like Terraform rely on a separate state file that is then your code.

So your code in your Git is not the real code anymore. And you really need to, you really need to play that catch up game with your, between your state and reality. Our approach is totally different.

What we say is, first of all, we embrace, embrace the reality. The reality looks like everybody will be using their own tool. You cannot enforce in a, a reasonably well in any reasonable setup.

You cannot enforce one single tool. So let's forget about that. Now, what do you want to do?

You want to give them their own tools, let them move of their, at their own speed, as we already discussed, but then catch them with a single source of truth. And that is the most important thing. That is what FormIt does.

Forme can perfectly coexist with any other ISC tool, click ops, any other security tool that makes changes in your cloud, including your own cloud providers that can constantly make changes. It is an agent-based tool that is important. So it's not what you run exclusively on your local machine and babysit it?

No, you don't have to. It has an active backend component, and this active backing component takes over the responsibility of actually converging the work to the proper result. And, uh, as I said, it's code in, code out.

This means that you can make changes in this platform at any granularity for any resource in your infrastructure. Um, with as minimum, um, blast radius as you want, you can do whole rollouts. You can reconcile a hundred percent from your code.

You can consider us as your source of truth regarding the code because you are extracting resources from their current real state. You are extracting them as code, you're changing them, you're putting them back, and we apply the change. So you always work through code, uh, with this tool.

And we're catching up by automated, uh, fully automated discovery and synchronization, we're catching up on what's happening in your cloud estate. So we basically are absorbing any change that happens through whatever tool you use, whatever people are using, and version them all of the changes and put them into our internal database, which allows us also to completely eliminate, um, the necessity for tools like Git. Well, let's say in the future, because when we early, we're still developing that, but what you already get is a fully versioned a hundred percent code oriented tool.

Charlie, couple questions. Sure. First of all, you mentioned of course, Terraform ha, HashiCorp, you know, our audience is sophisticated on this.

They know Linux Foundation or CNCF, you know, uh, open Tofu Yes. Is the fork. Sure.

But it, you know, it's in that same vein as as as Terraform, though they are, they're diverging. Right. As time goes on, I think you'll see more of a divergence between the two of 'em.

Um, the other big thing with platform engineering is it's become so much about internal developer platforms, backstage, things like this. Um, where almost that has become, if you will, the mission, right, maintaining the IDEI, I think there's a bigger mission for platform engineering than just the id. I don't want to minimize, right.

The id, you know, they, having an an ID is very important. Yeah. But there's, there's more to the platform than the ID is I guess what I'm saying.

Right. Um, now your tool for a, for a is also open source. Yes, sir.

Correct. Yep. And you guys are maintaining that.

What about the community? Yeah, we're gonna build a community. I mean, everything in the system is a plugin, basically.

So it's so pluggable that people can develop their own plugins, they can maintain their own licenses for the plugins, of course. Uh, as, uh, a uh, company that is funded by, uh, investors and has a commercial meaning we are going to go after well monetary aspects of the business. Um, but right now it is important to us to open this to the world and, uh, allow everybody to enjoy the infrastructures code it as it was supposed to be.

That is the most important thing for us right now. But of course, we're welcoming contributions. I, we want to build a healthy ecosystem.

We want to build a marketplace, we want to build all of these things. They're coming. Definitely the important, uh, aspect, uh, regarding this, uh, um, developer, uh, portals that you mentioned.

Well, the problem they all have is that in the space of infrastructure, they have to opt out to what is there. So this means that some of them are ending up generating Terraform code. And the problem with generation of Terraform code like HCL code, somebody needs to check it still.

I mean, you cannot rely on, on uh, uh, machines, uh, no spitting out proper code. Um, no, not even ai. And, uh, this is number one.

The number two is that, you know, even if you have that developer portal, your infrastructure management is still not solved. You want to have it on proper feet. And, uh, the abstractions through the portal only means that you need to provide services that are reusable and, uh, um, can be claimed by, if you wish, by those developers.

But in order to create an infrastructure service, you need to think, when you think about a database, you think about a t-shirt size of a database. You don't talk about gigabytes whatsoever. Not a single developer knows how many gigabytes this database will ever need.

Who knows that? Nobody knows that this is not my job as a developer. So the platform team needs to create abstractions.

And guess what? You can't create abstractions with the current tools. It is literally impossible.

That's what we give them as an add-on. You can actually create layers of abstraction involving every single engineer in your team at the level of their experience and in the level of their responsibility. This is another important aspect.

Excellent. Agreed. Um, I, besides the platform dial engineering, I'd imagine formula is also on like GitHub and so forth.

Yes, absolutely. You wouldn't happen to know the GitHub, uh, URL, would you? Yeah, it's, uh, so the company is called Platform Engineering Lab.

So GitHub is platform engineering Lab slash for me. Excellent. I just wanna make sure we give 'em everything.

Yes, Absolutely. Um, what about Discord or Slack or other ways for the community to communicate? Yes, there's a Discord.

The Discord is in the Read me on GitHub, the Discord is in the Read Me on the organization. That's all set up. So people are welcome to join and there are GitHub discussions.

We're completely open and, uh, actually very excited about talking to as many users as possible. I love it. Pavo, we're about outta time, but I wanna wish you all the success.

Thank You so much. On, on this new venture platform, engineering labs. Hey, we're always looking for new content and exciting things.

com, so keep us posted and you know, we've got a good place to put some news up there. You got It. So we Thank I hope to hear plenty from you.

Alright, thank you. Pavlo Baron, co-founder and CEO for platform engineering labs. engineering, uh, makers of for, uh, open source infrastructure is CLO code platform built for the future.

You're watching Text Drunk tv. We'll be right back.