env zero Connects IaC Governance to Cloud Intelligence
Alan Shimel talks with Oved Lourie, CTO of env zero, at PlatformCon 2026 about how platform teams can simplify infrastructure delivery while keeping governance, policy and drift under control. Lourie explains how env zero supports Terraform, OpenTofu and self-service infrastructure workflows, while CloudQuery adds an intelligence layer that extracts cloud data and maps relationships so AI agents can reason over accurate infrastructure context. The conversation also covers platform engineering, infrastructure ontology, developer self-service, deterministic AI decisions, cloud data complexity and why reliable infrastructure data is essential as autonomous systems take on more operational responsibility.
Transcript
Hey everyone, we're back here at PlatformCon. I wish I could show you the other side of the camera here, because we're surrounded by windows and if I look this way, I'm looking east at the Empire State Building. This way is north at the pencil buildings on 57th Street.
And down here, we've got the, well it's not the World Trade Center, but Freedom Tower, a World Financial Center downtown. It's like a multimillion-dollar view. It's incredible.
Yeah. If you could build an apartment here, I don't know what you would charge for it. Oh.
Anyway, my next guest is Oved Lurie. Oved is the CEO? CTO.
CTO, excuse me. Yes. Of End- But thank you.
Maybe, env0. E-N-V-Z-E-R-O. First of all, Oved, welcome to Techstrong TV.
Thank you. It's great to have you on here. Let's start with a little bit about you.
All right. Give us a sense of where you've been, how you got here. Yeah.
So, been in the space for about 25, 30 years now. I would always say my foundational years I spent in IT, and probably the most significant thing I did that proves that I can talk the talk is I skipped the first Red Sox World Series in 2003. Really?
I was a season ticket holder, so that I could go to the first VMworld instead. No. Yes, because I was like- Oh, you're crazy ...
this thing is... No, because I was incredibly lazy, not crazy. Oh, okay.
" Ah. This thing allows me to just take servers as files, ship them over somewhere. No more hardware, no more LUNs.
It was so simple, but I couldn't figure out how to do it in 2003, and I was like, "I'm going to go there, and I'm going to get all those engineers to tell me all the secrets" and that's what I did. " It was incredible, and that changed the world as we saw it. I remember, the snap.
You could do like a snapshot and have your- Take the snapshot, take the file. It's like an MP3. Ship it over, make your life so simple.
Yeah. And I remember in 2007, Gartner told us we were over-virtualized. Nobody knew where Wintel was going to go, and we should probably start buying servers again.
" But- Well, that's Gartner for you. Yeah. So a lot of time in IT, 10 years.
I saw XenServer get acquired by Citrix. Spent some time there thinking I could take on VMware, but as we all know, that was not going to happen. You had to wait till now.
Yeah. And then, for the last 15 years, spent time at Turbonomic, which was acquired by IBM in 2021, and had a number of different roles. But, in the last few years at IBM, I ran worldwide tech sales for IT automation.
Really? Yeah. Oh, very cool.
So tell me how you came to env0. So, a few of the folks that I'd worked with at Turbonomic, the CEO, Steve Corndell, him and I worked together for a really long time, and we'd been talking about this for quite a while. And earlier this year, there was a merger between CloudQuery and env0 that made things super interesting, to the point that it became the time to come join the team and lead the R&D and really help us get to that next level.
Excellent. So look, a lot of people in our audience may not be familiar with env0. Yep.
Give us that story. Yeah. So env0, it's quite an incredible company.
Our customers don't just start with env0 and then stall. They start with it, they buy more, they continue to use it, and it works in the background because fundamentally, the platform is there to enable people to have easy access and seamless access to run their codes and their apps the way they want them to run. And so we're trying to make it easier for developers to have the best possible experience, where they don't have to keep going to the platform team every time they want something, and the platform team doesn't have to be interrupted.
So we allow infrastructure as code, whether it's Terraform or OpenTofu. We make it so simple for developers to only have access to what they need, to not even know what infrastructure is, to simply take their code and deploy it seamlessly through a self-service model and just make it an easy world for everybody to continuously get infrastructure deployed without even knowing it's there. Look, forgive me, but it sounds almost too good to be true, right?
It's incredible. It really is. Terraform gives you the code to deploy it.
env0 just gives you an incredible way, not just to deploy it, but how do I apply governance to it? Because I need to make sure everybody's... How do I apply the right policies to it?
How do I make sure I'm not drifting from the intended state that I deployed things into? And then when we get into CloudQuery, it gets super interesting on top of that. We're going to talk about CloudQuery in a second.
I want to stick to the env0 piece of it here in terms of, look, we're here at PlatformCon. Yep. People are using platforms.
They're using IDPs, they're using open source IDP, commercial IDP. We talk about golden paths, all of these things. Now, of course, everybody's talking about agentics and AI.
Yep. Where does, in that stack, if you will, where does env0 sit? So it's above the IDP.
It's basically, you're creating your Terraform modules, you're creating your templates, you've got your code, you want to deploy it. That's where we sit. Okay.
It's at that time where you're either continuously deploying, where it does it automatically, or a developer is deciding it's time to roll out this code and deploy it on infrastructure, continuously maintain it. So that's really where we sit on top of that. I love it.
And so does it make a difference what platform we've got underneath? No. It doesn't really matter.
We have customers with all flavors sitting underneath us. And same thing for the infrastructure as code. If you've got open source Terraform, if you've got OpenTofu, doesn't really matter to us.
Really. Excellent. You mentioned Cloud Query, right?
Recent news around that. Yep. So Cloud Query, why it was so interesting is if you think about the ADP- Mm-hmm ...
what ends up happening is you've always had people in the loop who, when there's a problem, when something's not right, they can always work around it. But now you have agents that are reasoning over data, and if that data's wrong, it's not working around it, it's taking it as truth, and it's acting on it. And so as we shift into this world where agents are really so far in the loop, hitting your APIs, hitting your data all the time, and trying to pull it from all the sources, whether it's in the software itself, the definition's there, whether it's in the intended code, it's got to be right if you're going to close this loop.
And so what Cloud Query did was they built this incredible technology, and it might sound really simple, to extract your cloud data and- I know this is not simple though because I've lived that nightmare You look at AWS, for instance. You've got over 1,500 APIs just for the services that are continuously changing. Then you've got Google, then you've got Azure, then you've got ServiceNow, then you've got Wiz.
Then you've got all these different systems, so it's not just the data itself, it's how is the data related? Who owns that data? Because if you get it wrong and you make the wrong decision, something's going to go wrong.
Right. So what Cloud Query has done has made it incredibly easy to extract this data and put it in a place where AI can now reason over this data and be right. You've got to have the right ontology.
You've got to have the right relationships in place- I agree ... so that when the decision is made, what do you do with it? It's not just about, okay, now I've got this piece where I've got the right decision, and now I've got this piece that can do the action.
How do I tie this together? And I think we can get more into Drift and what that really is. We just went through a migration, I'm still in the throes of it, to tell you the truth, moving to a headless CMS from regular WordPress.
And you would think that that's an easy thing. Oh. No.
A nightmare because it's not just the data. You're 100% right, the data was easy. Yes.
That would probably be the easiest thing. It was all the plug-in information and the configuration and just the scripting. Little things like you don't think about.
What were the category tags? What were the different tags for these things? The image libraries, man.
Yeah. A nightmare. Yeah.
I talk to customers all the time, and when you start to get into ownership, it's the hardest thing to figure out because I can define it in my software. I can tell you what I want to deploy. I can go deploy it.
But as soon as that gets into the environment, number one, it's going to change, and I'm going to have to figure out who owns it. But number two, drift isn't just drift from what I intended my code to be in the deployment. Drift is drift from the intended state of the cloud.
At any point in time, a vulnerability can appear. It's like dipping your toe in the water. I've drifted from secure.
At any point in time, I can drift from a million different things. It's not just drift from my code, and so I have to reason over it. And then what- And AI itself drifts.
AI itself, if it doesn't have a graph- It's non-deterministic. Right ... Right.
And that's what your cloud is. It's actually data. It's not infrastructure, it's data, and the only way that it can become deterministic is if you have an ontology that defines those relationships for the AI to reason over- But it's almost like quantum, right?
It's- Because as soon as you do that, you blew it all the- And that's why we're starting in this very narrow wedge, which is really around this infrastructure because that's the piece that we can control. Everybody's talking about AI and reasoning and agents and data and orchestration. Problem is, even if you have good reasoning, even if you have the best decisions in the world, what are you doing next?
Are you handing it off to another system? And now you're three hops away from your detection, which means that that's where the blast happens. Mm-hmm.
So it stands to reason that you actually want to tightly couple the decision to the action. And since I've got M0, which already knows where the code is stored, already understands what your deployments are, can easily take this reasoning and now handle this continuous reasoning of itself in the cloud. That's where we're going, where the cloud understands itself and makes decisions to keep you in that state all the time and takes those actions autonomously.
So I don't have to ask you then how you're working with Cloud Query because I think you just explained it. Yes. This is the best of both worlds because I know how to deploy, and I also know what your current state is of the cloud at any given time, instead of just looking at some of the metadata that you've been collecting from a bunch of tools.
It stands to reason that's the best way to know what the current state is. Not at 3:00 in the morning trying to figure out which service is the one that, and who owns that service- Exactly ... because something has happened.
Right. That's impossible. The new metric that everyone should be concerned about is the time to know.
We think about MTTR. Right. MTTR is easy.
The time to recover is easy once I understand, but before I even understand- You got to know ... I have to detect. I got to know.
I got to know exactly what it is that I'm going after Absolutely. Otherwise, you're just shooting in the dark anyway. Yeah.
Let me ask a business question. Yep. Do you see the future of CloudQuery and env0 as two separate entities, two separate names, or does this come together as one solution?
I think we will not have it as two separate names. It will come together as one solution, but it doesn't come together as a CloudQuery dashboard. Everybody's got all the dashboards in the world.
Everybody's got a million tools. I think it comes together as this intelligence layer because it's able to reason deterministically because it understands what relationships are, it understands what the taxonomy and what the actual definition of entities are, so that you don't have to look at it. You can define policy, you can define guardrails.
You can even figure those out on its own. A desired state versus a rule, they're two completely different things. But I think we start with the rules to apply to the cloud, and it becomes a part of env0.
But it also becomes something that is able to make these decisions, and you don't have to shift immediately, but we will have a very tightly coupled integration between the two systems. You know what, we talking all this time, I don't think we mentioned the website. The website?
Yeah. People want to get more information. Where do we go?
com. Go to- E-N-V 0 ... env0.
Easy does it. io, so we still have two different websites. We actually have a free trial, and we have an easy entry point into CloudQuery, so if you just want to get familiar with natural language across your cloud, it's pretty awesome because if you have a number of cloud accounts across multiple clouds, projects, accounts, it's actually hard to get that data in.
It's hard to maintain those APIs, but you wouldn't believe how many people can't even tell you how many storage buckets they have that are unencrypted and internet-facing within three seconds. m. to figure out what it is.
It should be right at your fingertips. m. or on weekends.
Exactly. Right. And when it happens, you spend hours, days, weeks with just projects.
So CloudQuery, got a trial out there. We've got the old CLA, env0. We have a trial out there for that as well, but really easy to get started and our teams are incredibly strong and will help get you up and running.
I love it. Oved, this is your first time on Techstrong TV. First time ever speaking to people.
Really? Yeah. I would have never known.
Just kidding. Thank you. But no, seriously, come back.
We do this every day. Oh, this is fantastic. We do three hours of video a day.
If you're not in person, we do them on virtual, whatever platforms we're using that day. Thank you for having me. Please come back.
I appreciate it. All right. Oved Lurie here for env0 and CloudQuery- Yes ...
dot io, I should mention. We're going to be back with more of our PlatformCon coverage. Stay tuned.