Yesterday I was a Senior Manager at ezCater, reporting to the Director of Engineering. Today I’m sitting at my desk at 6am, three cups of coffee in, building the most ambitious thing I’ve ever attempted.
I got laid off on November 20th.
And I’ve never been more certain about anything in my life.
The Thing I’ve Always Known
There’s a peculiar kind of clarity that comes from being forced to stop performing. For the last few months, I felt it building—a low hum of misalignment. I was doing well at management. I got promoted. I had autonomy over people, budgets, and strategy. By any external metric, I was succeeding.
But I was slowly suffocating.
This wasn’t burnout. It wasn’t the job. It was simpler: I’m not built for managing people. I’m built for building systems. I’ve managed for a decade, and I did it well. But well isn’t the same as whole.
ZenithMind OS told me something I needed to hear in a way my rational mind finally couldn’t argue with:
Your psychology is built for building, not managing. Your core pattern (INTJ + Enneagram 5w6 + Kolbe 7-3-6-2) makes you happiest when solving systems, energized when creating frameworks, at your best when designing machines—not running them.
The moment I read that, something clicked. Not the “oh, interesting observation” kind of click. The I’ve been pretending for five years kind of click.
Management requires constant context-switching, emotional presence, interpersonal maintenance—things that drain rather than charge my battery. I can perform management at a high level. I’ve done it for decades. But it’s not where my identity refuels.
My identity refuels when I’m architecting, building, debugging, evolving.
The Pattern That Won’t Quit
I’m not new to this. Building is my oldest, truest pattern.
In the late ’80s, I wrote Fantasy Football software—alone, late into the night until 1am, then straight into my day job. Months of this. I wasn’t trying to get rich. I wasn’t trying to impress anyone. I was solving a problem that existed in my mind, turning thoughts into artifacts, making the internal external.
That feeling never left me. It just got buried under mortgages and career ladders.
By May 2024, I couldn’t ignore it anymore. I started exploring what it meant to build adjacent to AI—learning, experimenting, thinking at depth about how to architect with these new tools. The tools weren’t quite there yet, but the possibility was.
Then Fall 2024 arrived, and everything changed. Cursor showed up. And suddenly, building wasn’t theoretical anymore—it was possible. Real. When Claude Code, Codex, and Gemini arrived in the pipeline, it was rocket fuel on something already burning.
This wasn’t a side hustle experiment—it was an identity emergency. For the first time in years, I felt alive when I coded. Not exhausted. Not playing a role. Not managing up and coaching down.
Alive.
And ironic as it sounds: building with AI—”vibe coding”—feels like playing a game. Except when you win, you have a real product. A real system. A real artifact that reflects the clarity in your mind.
The Operating System Shift
I’m approaching 2026 with a completely different architecture than most builders I know. And I didn’t invent it—I’m standing on the shoulders of people who’ve been thinking about this for years.
AI has made generating ideas trivial. That’s actually the problem now, not the solution. The constraint isn’t ideation—it’s speed of qualification and validation.
Codie Sanchez has been hammering on this: in a world where everyone has access to the same AI tools, the only real moat left is speed. Compressing the OODA loop. This is the core thesis of Startup Empire—compress how fast you can test, iterate, and kill ideas.
So instead of betting everything on a single product, I’m building an Idea Orchestrator—a system that:
- Ingests researched opportunities from tools like Ideabrowser, founder networks, and domain research
- Runs them through a qualification filter I’ve built:
- Agent-first workflows (can this be automated?)
- Vertical focus (is this a real problem?)
- Rich adjacency (do I have unfair advantage here?)
- 24-hour testability (can I validate this fast?)
- Pride test (would I be proud to be “the X-guy” years from now?)
- Pushes the best ideas into rapid-fire experiments before touching code
- Tests through real conversations, concierge pilots, pre-selling—long before I commit
This pulls from:
- Greg Isenberg’s Audience → Community → Product model (the only sustainable path)
- Alex Hormozi’s value equation (find desperate problems, high purchasing power)
- Codie Sanchez’s 2026 operator rules (sell before you’re ready, build distribution early)
- My own manifesto on staying in lanes and letting AI amplify what you’re already good at
The goal isn’t to guess right. It’s to test fast, kill faster, and let winners emerge from the noise.
By treating ideas like a supply chain instead of a single shot, I can cast multiple fishing lines without drowning in complexity. The orchestration system becomes the engine. My build logs and public work become the distribution flywheel.
Over time, this should surface one or two vertical opportunities with real traction—enough to build a portfolio of durable AI-powered businesses with real runway and real impact.
Why This Matters (And Why You Should Watch)
I’m not unique. But I’m intentional in a way that is rare.
Most builders either:
- Spin their wheels optimizing for someone else’s roadmap
- Chase trends instead of solving real problems
- Move too slow because they’re waiting for perfect clarity
- Build alone without distribution or audience
I’m deliberately constructing something different:
- Autonomous building without organizational drag
- Public iteration so the world can see what works (and what doesn’t)
- Speed as a moat—testing ideas in days, not quarters
- Distribution from day one—vibeengineer.ing isn’t a portfolio site; it’s the flywheel
- AI-augmented workflows—not AI-dependent, but AI-amplified
The experiment is simple: build in public, move with unreasonable speed, and let the data—not the dream—decide what gets built.
If it works, you’re looking at the blueprint for a new operating system around AI-powered SaaS and agent businesses.
If it doesn’t, you get to watch someone learn in real-time, transparently, at scale.
Either way, there’s no bullshit here. Just builds, logs, and honest iteration.
What’s Next
The plan is straightforward:
- Finalize the Idea Orchestrator architecture
- Document the qualification filter with real examples
- Publish the research framework for vertical selection
- Set up the experimental cadence (rapid testing, daily logs)
- Run the first ideas through the qualification funnel
- Conduct real pre-sell conversations and gather customer feedback
- Ship the first product that makes real money
I’m not moving fast because I’m desperate. I’m moving fast because I’m finally doing what I was built to do.
If you’re interested in watching this unfold—the wins, the failures, the honest rebuilds—you’re in the right place.
This is vibeengineer.ing. This is what happens when someone stops managing other people’s work and starts architecting their own future.
Welcome to the build.
📋 Disclosure: This article contains affiliate links to ZenithMind OS, Startup Empire, Kit, and Postmark. We only recommend products and services we’ve genuinely tested and believe provide value. See our Terms of Service for details.