I Stopped Outsourcing My Ideas
There's a moment — and if you've been in tech long enough, you know exactly what I'm talking about — where you stop building and start managing. You trade the terminal for spreadsheets. You swap debugging sessions for client calls. And slowly, without really noticing, you become someone who talks about software instead of someone who makes it.
That was me for the past few years. Running Netspace, our digital agency, meant I had to focus on the bigger picture. Strategy, hiring, client relationships, operations. I'm an engineer by training, I know how long it takes to ship a real product. The bugs, the scope creep, the gap between what you imagined and what actually lands in production. So I made my peace with it. I let go of the code.
Until Claude changed everything.
The moment it clicked
I'd tried AI coding assistants before. They were nice. Helpful for boilerplate, autocomplete on steroids. But Claude Opus was something else entirely. The first time I used it seriously — not for a toy project, but for something I actually needed — I realized the rules had changed.
I wasn't just getting code suggestions. I was having a conversation with something that understood architecture, tradeoffs, context. I could describe what I wanted, iterate on it in real time, and watch a working product take shape in hours instead of weeks.
As an engineer, that hit different. Because I knew exactly how much work was being compressed. I knew what it meant to go from idea to working prototype in an afternoon. This wasn't automation. This was amplification.
Remindian — building for myself again
The first real thing I built was Remindian. Not a client project, not an agency deliverable — something for me.
I live in Obsidian. All my tasks, notes, everything runs through it. But there's one problem: I can't add tasks through Siri, and I can't access them when I'm away from my Mac. Apple Reminders can do all of that, but it doesn't talk to Obsidian.
So I built the bridge. A native Mac app that syncs Obsidian tasks with Apple Reminders. Two-way. Real-time.
I shipped it open source. The community picked it up. People started using it, filing issues, requesting features. I iterated. v2, v3, v4. Each version better than the last, each one built faster than any traditional dev cycle would have allowed.
But here's the thing that mattered most: I got my confidence back. That feeling of making something — of solving your own problem with your own hands — I'd forgotten how good that was.
Killing the SaaS dependency
Once Remindian worked, a switch flipped. I started looking at every tool we were paying for at Netspace and asking: do we actually need this, or can we build something better?
BannerBear was generating our social media visuals. It worked, but it was generic, limited, and expensive for what it did. So I built Suwar — our own image generation platform, tailored exactly to our workflow and our clients' brands. No more compromises, no more "the tool doesn't support that."
Mailchimp was handling our email campaigns. Again — it worked, but we were paying for features we didn't use while missing the ones we needed. So I built Barq — lean, fast, does exactly what we need, nothing more.
These aren't side projects. These are production tools that run our agency every day. And the fact that I could build them — as someone who hadn't written serious code in years — says everything about where we are right now.
These apps weren't born from vibe coding
I want to be precise about this, because it matters.
Remindian, Suwar, Barq — none of them were born from vibe coding. They were born from real problems, real frustrations, real needs. What vibe coding did was remove the wall between having the idea and building it. It put me back on the path of development after years of being away from it.
The ideas were always there. The engineering instinct was always there. What was missing was the bridge — a way to go from "I know exactly what this should do" to "here it is, running" without the months of development overhead that made personal projects impossible when you're already running a company.
What this means if you're reading this
If you're an entrepreneur, an agency owner, a technical founder who drifted away from building — you probably recognize yourself in this story. You have a backlog of ideas you never shipped. Tools you wish existed. Problems you've been working around instead of solving.
The barrier is gone. Not lowered — gone.
You don't need to hire a dev team. You don't need to learn a new framework. You don't need three months and a budget. You need your domain expertise, a clear idea of what you want, and a conversation with the right AI.
That's it. That's the whole secret.
And if you want someone to show you how — that's exactly what I do. I coach entrepreneurs and teams to go from idea to product using these tools. Not because the AI does everything for you, but because your knowledge is the most valuable part of the equation. The AI just removes the friction.
The real shift
People talk about AI replacing developers. That's not what I've experienced. What I've experienced is AI giving builders their superpowers back. The ones they lost to management, to delegation, to the natural evolution of running a business.
I'm still the CEO of Netspace. I still manage clients, strategy, operations. But now I also build. And that changes everything — not just for me, but for what we can offer, how fast we can move, and what becomes possible when your bottleneck isn't development anymore.
The best product ideas come from people who deeply understand the problem. If that's you, stop waiting. Start building.