Cielara is going to KubeCon + CloudNativeCon Europe 2026 in Amsterdam, March 23–26, and we couldn't be more excited. If you're a developer tired of firefighting outages caused by misconfigurations, drift, and blind deployments, we'd love to meet you.

Why KubeCon Matters to Us

There's something special about KubeCon. It's where the people actually building and running cloud-native systems get together, swap war stories, and figure out what's next. That energy is exactly why we started Cielara in the first place.

Here's what we keep hearing from teams: Kubernetes adoption is growing fast, AI workloads are hitting production, multi-cloud is the reality, and the complexity is outpacing everyone's ability to reason about what's actually happening. Changes ship quickly, but nobody has the full picture of what those changes will do.

We're building the thing that gives you that picture.

So, What Exactly Is Cielara?

Think of it as a Production World Model. A live, always-up-to-date context graph of your entire cloud and AI infrastructure. We map every resource, dependency, and configuration across your environment (cloud, AI, CI/CD, IaC, Git — all of it) into a single model your team can actually see and query and your coding agent can actually reason about.

It's not another dashboard full of alerts. Cielara shows you what's going to break, why, and what you can do about it — before you hit deploy.

Under the hood, the platform gives you a live infrastructure model, pre-deployment simulation, an ontology graph that learns how your org actually works, built-in policy enforcement, intelligent remediation, and the ability to ask your infrastructure questions in plain English. Cielara is like talking to the one teammate who somehow knows how and where everything is connected.

What We'll Be Showing in Amsterdam

Swing by our booth 1261, and we'll walk you through the real thing. Here's a taste of what you'll see:

• Catch risky changes before production — simulate the blast radius before a single line of code ships
• Spot environment drift the moment it happens — no more wondering if live infra still matches your IaC
• Intercept AI-generated misconfigurations — because AI-written code and config changes are great until they break something at 2 AM
• Ask your infrastructure anything — "where are risks piling up?" or "what's downstream of this service?" — and get real answers grounded in your actual environment

Teams already using Cielara are seeing 60% fewer failed deployments, 75% fewer outages, and 80% faster resolution times. Those aren't projections — that's what's happening today.

We Built This Because We Lived It

I was on-call at a Fortune 500 company a few years back during Black Friday. A team had added a caching layer to improve performance of their database service. The code was nearly perfect and ran great under regular load. However, the developer forgot to put an upper bound on memory usage. When the holiday traffic surge hit, the service buckled and took down the entire retailer's website. Nobody had a way to see that coming. Now we do.

That's the moment Cielara was born. A chain of small, disconnected misconfigurations that nobody could fully trace. Systems that weren't fragile — just too complex for anyone to see the whole picture. Our founders came from engineering leadership roles at Uber, Microsoft Co-Pilot, and HPE, and they'd had enough of explaining outages after the fact. So they set out to build a way to prevent them.

Our research team includes some of the leading minds in Graph Neural Networks and Graph RAG, and we're lucky to have advisors from Microsoft Research, Hugging Face, OpenAI, and Google DeepMind backing the work.

Let's Grab a Coffee in Amsterdam

Whether you're wrangling Kubernetes across multiple clouds, shipping LLMs into production, or just hoping for one week without a 3 AM pager alert — come find us. We'd love to hear what you're dealing with and show you how Cielara could help.

See you in Amsterdam!

Book a time to chat at KubeCon EU: Cielara.AI

About the Authors

Max is the Lead of Agentic Engineering at Causal Dynamics Lab. At Microsoft, he spearheaded the rollout of thousands of agentic actions across Microsoft Copilot and Copilot Studio to a user base of 400 million. His work sits at the intersection of agentic engineering, neuromorphic NLP, and on-device language models. He's a three-time founder with a Master's in Engineering from ASU and an MBA from Foster. Find him at https://www.linkedin.com/in/mmurshed/

Bowen Zhu is the Lead of the World Model Foundation at Causal Dynamics Lab, where he leads foundational engineering and AI systems research. Previously, he held engineering leadership roles at Uber and Microsoft, building large-scale platform, security, and AI systems that serve millions of users and developers. He was an early pioneer of AI-driven intent-based access management, advancing systems that understand context and intent rather than relying solely on static rules. His current work focuses on agent memory and enterprise AI systems that can reason, adapt, and improve from enterprise wide context. He believes the future of AI will be defined not only by model intelligence, but by the systems that enable agents to learn, remember, and operate reliably at scale. Find him at https://www.linkedin.com/in/bowen-zhu-6a724872/