- Prague-based PaaS Zerops received USD 2M of Seed funding to expand cloud infrastructure globally
- The company’s goal is to eliminate development-production separation challenges
- Its proprietary Zerops Control Panel (ZCP) connects AI agents with real cloud environments directly
- Zerops’ philosophy focuses on predictable infrastructure rather than unnecessary complexity
This May, Zerops—the renowned Czech Platform-as-a-Service startup redesigning cloud architecture—raised its USD 2M seed round from the startup’s standing supporter, the well-known VC fund Gi21 Capital (invested in Ranketta).
A New Chapter of Growth
The company’s story had already been unusual. It was never a typical startup born in a garage with a clean cap table and a straightforward path from idea to funding. Instead, Zerops had spent years being built inside vshosting, survived an acquisition, was effectively abandoned, and was later bought back by the team that had created it.
Last time ITKeyMedia spoke with Zerops was in summer 2024. At the time, the company had just secured its Pre-Seed investment and was entering a new chapter. The immediate goal was not to compete head-on with heavily funded cloud platforms, but to prove that a small team with deep infrastructure expertise could build something developers genuinely wanted to use.
Since then, the company has moved from rebuilding its foundations to challenging one of the biggest assumptions in modern cloud development: that development and production environments should be separate.
Now that Zerops raised its new USD 2M Seed round from Gi21 Capital, the new funding should support international infrastructure expansion, including the US and Asia, further product development, and team growth.
Aleš Rechtorík, Co-Founder and CEO at Zerops
But according to Zerops’ co-founder and CEO Aleš Rechtorík, the biggest change over the past year has not been the funding itself. It has been a shift in how the industry views the problem Zerops has been trying to solve from the beginning.
Turning Expert Infrastructure into an Automated Platform
The original idea behind Zerops came from vshosting, where the team managed infrastructure for demanding customers, including large e-commerce platforms and high-traffic applications. The logic was simple: if experts could build reliable infrastructure setups manually, why not turn that knowledge into an automated platform? That became Zerops — a play on DevOps, meaning that developers should not need an entire team of infrastructure specialists just to run their applications.
However, the company quickly discovered that removing operational complexity is much harder than simply hiding it. While many platforms promise simplicity, it often comes from merely abstracting away infrastructure details. This works until something breaks and developers need to understand what is actually happening underneath.
Zerops’ goal, instead, is not to hide infrastructure. It is to remove unnecessary decisions while keeping the visibility developers need. The core idea behind Zerops today is that development and production should not be two separate worlds.
Eliminating the Gap Between Development and Production
Traditionally, developers create applications in one environment, then deploy them into another environment with different configurations, dependencies, permissions and infrastructure, — which works only sometimes. Should it fail, teams need to spend hours trying to understand why something that worked yesterday suddenly misbehaves after deployment.
‘Most platforms ask you to trust that development and production are close enough. We removed the gap entirely by rethinking how cloud architecture should work from the ground up. That’s the same guarantee we now give AI coding agents, and it’s why the code they produce is production-ready from the first deployment. What works once continues to work as applications scale. Our goal is to make running software predictable, not something teams have to constantly debug,’ Mr Rechtorík states.
According to Zerops, the biggest misunderstanding around identical development and production environments is that people think it is simply about containers. The team believes this is actually the easy part. The difficult part is making sure everything surrounding an application behaves the same as well: networking, databases, permissions, storage, deployments, background services, scaling. That’s the source of most surprises.
As such, instead of making production resemble development, Zerops tries to eliminate the conceptual difference between the two. The environment remains the same. The scale changes, but the foundation does not.
AI Agents Reveal the Old Infrastructure Problem
The timing of Zerops’ approach is closely connected to the rise of AI coding tools. The company argues that AI agents are not creating an entirely new infrastructure problem, but exposing an existing one.
It is known how experienced developers often compensated for imperfect environments through years of accumulated knowledge. They knew which assumptions were dangerous, which configurations often failed and where to look when something breaks.
AI agents don’t enjoy this kind of context. As the development-versus-production gap has existed for decades, AI agents simply hit that wall much faster because they don’t know how to improvise around broken environments the way experienced engineers do , — this is the line of thought behind Zerops Control Panel (ZCP), a feature designed around AI-driven development.
Rather than letting AI agents operate in isolated environments, ZCP connects them directly to real cloud infrastructure inside a Zerops project, allowing them to build, deploy, and debug applications in conditions close to actual production.
The idea is not to replace developers. Instead AI agents are more like incredibly fast interns.
Keeping Infrastructure Transparent Without Overwhelming Developers
One of the central questions facing cloud platforms is how much infrastructure developers should actually see: too much complexity creates unnecessary work, and too much abstraction creates blind spots. Apparently, Zerops has chosen a middle ground: infrastructure should become easier, but not mysterious.
The analogy is similar to driving a modern car: Drivers don’t need to build the engine, but they still expect a dashboard showing what’s happening when something goes wrong.
With AI agents becoming part of software development workflows, transparency like this becomes even more important. Both humans and AI agents need access to reliable information about the systems they operate.
‘The market is reaching an inflection point. Rising cloud costs are forcing a shift, while AI is changing not just how software is written, but who is building and running it. We’re moving from millions of developers to millions of developers working alongside AI agents. Most platforms weren’t designed for either of these changes. Zerops was. Its economics come from owning the full stack, and its architecture works because it never abstracts away the underlying infrastructure,’ Gi21 Capital’ managing partner Damir Špoljarič comments.
The Economics of Expanding Beyond Europe
Damir Špoljarič, Managing Partner at Gi21 Capital
The new funding will allow Zerops to expand its infrastructure into North America and Asia, including a new data center in Singapore. Here, international expansion introduces a challenge familiar to every infrastructure company: geography increases complexity and cost.
The apparent answer lies in controlling more of the stack: owning and optimizing more of Zerops’ infrastructure will allow for maintaining cost advantages while expanding closer to customers.
However, the more important question would be whether developers in new markets see Zerops as a platform they can build on: Are developers deploying real production systems? Are they staying when their applications grow? Are they recommending the platform to others? These are the signals for Zerops to watch.
From Underdog to a Bigger Ambition
If a year ago, Zerops was an underdog with a strong product and a small team trying to catch up with competitors that had raised significantly more money, today, the company is still the underdog — but with a much bigger ambition.
The original promise remains the same: developers should spend their time building software, not fighting infrastructure. The difference is that the problem has become larger. As AI changes how software is created, cloud infrastructure has to evolve as well. From this point of view, the future is not about making infrastructure disappear, but about making infrastructure predictable.
The Future of Cloud Infrastructure: Predictability over Complexity
In an increasingly AI-driven software landscape, the ability to provide reliable, transparent, and scalable infrastructure may become one of the key differentiators between platforms that merely host applications and those that truly enable developers. Zerops’ approach addresses a fundamental challenge that has existed long before AI: reducing the friction between writing software and running it reliably in the real world. Whether the company can turn this vision into a globally adopted platform remains to be seen, but the bet is clear — the future of cloud infrastructure should be built around predictability, not complexity.
Kostiantyn is a freelance writer from Crimea but based in Lviv. He loves writing about IT and high tech because those topics are always upbeat and he’s an inherent optimist!
