Tomasz Cholewa
I didn't plan to become a Platform Strategist.I just couldn't stop looking for the real problem.
AI can now implement. What it can't do is walk into your organisation, read the room, and tell you what's actually going wrong. That's what I do — over two decades of platforms, with zero incentive to tell you what you want to hear.
From the basement to the stage
It started with a CD glued to a magazine. Red Hat 7.2. I was a teenager in Lublin, Poland, and I had no idea that installing Linux from that scratched disc would be the first domino in a chain reaction still going today.
Within months, I was running Linux as a shared internet router for my entire neighbourhood — not because someone asked me to, but because the technology was there and the problem was real.
My first job was as a Linux engineer, still an undergraduate. I automated everything I could touch — installations, configurations, provisioning. Not because my boss told me to. Because repetition felt like waste and manual steps meant friction: waiting, errors, dependency on whoever happened to know the procedure. I was looking for the bottleneck long before I had a word for it.
I moved to Warsaw. Started consulting. Helped Poland’s biggest bank, its largest telco, international financial institutions. Built OpenShift platforms. Taught Kubernetes to engineers across the country. Created a YouTube channel and a podcast under the name Cloudowski. Flew to the United States to present at Red Hat Summit.
Somewhere along the way, I stopped being the person who uses platforms and became the person who shapes them.
I advise at the strategy level because I've operated at every level below it.
— Tomasz CholewaThe disruption I didn’t see coming
Then AI arrived. Not the hype — the reality.
The training courses I’d built. The online content I’d spent years producing. The model of “teach engineers, they’ll pay” — disrupted, faster than I’d expected. It wasn’t a slow decline. It was a structural shift. The market moved and didn’t move back.
I sat with the question long enough to get an honest answer: what had I actually been doing all these years that AI couldn’t replace?
Not the technical knowledge. Something harder to name. The ability to walk into a room where smart people are disagreeing, see what’s actually in conflict, and say the thing no one has said yet. The judgment formed by watching the same decisions play out across dozens of organisations over two decades. The pattern recognition that doesn’t require a prompt.
That’s what I decided to build around. Not as another team of consultants. Not as a vendor. In a world where AI can generate a Kubernetes manifest in seconds, implementation is no longer the scarce resource. Judgment is.
Technology knowledge depreciates. Judgment compounds. You're not paying for what I know — you're paying for how I think.
— Tomasz CholewaHow I work now
I advise CTOs and VPs of Engineering as a thinking partner. Outside perspective. No implementation scope. No stake in the outcome.
I bring twenty years of pattern recognition to your specific decision. Not a framework. A point of view earned through consequence.
No vendors to protect, no partnerships to preserve, no implementation team to keep busy. When I tell you something doesn't make sense, there's nothing behind that opinion except the pattern I've recognised — no vendor to protect, no methodology to sell.
I'm most useful when you want a second opinion, not a validator. I'll tell you what I actually think — including when I think you're about to make a mistake.
I've seen the same decisions play out across dozens of organisations. I know which ones look right in the short term and produce expensive pain later.
What I believe
Any engineer can access any technical answer in seconds. What compounds, what remains genuinely scarce, is the judgment to know which knowledge applies, when, and at what cost.
They're building portals when they need governance. Adding tooling when they need clarity. Optimising for developer happiness metrics while structural debt accumulates quietly underneath.
Most platform adoption failures aren't caused by the wrong technology. They're caused by a platform that made things harder than doing them manually. The answer is rarely a new tool — it's removing the manual steps, tickets, and gates that make engineers work around the platform instead of through it. Everything as Code is not a methodology. It's the structural answer to friction.
Most advice comes with a hidden agenda — a vendor to protect, a methodology to sell, a partnership to preserve. I've spent twenty years watching this dynamic play out. My only loyalty is to the best decision for your situation.
If something here resonates, reach out. Tell me what you’re dealing with. I’ll tell you honestly whether I think I can help — and if I can’t, I’ll say that too, and point you somewhere better.
Or reach me directly: tomasz@cloudowski.com · LinkedIn