I Can't Code. I Made 4 Working Apps in 30 Days.

I’m not a developer. Never have been. But in the last 30 days, I made four working apps. A trading dashboard connected to a live CRM. A real estate report generator. A mood journaling app I’ve been dreaming about for a year and a half. And an AI agent deployment as a bonus.

No development team. No outsourcing. Just me and AI.

Why Most Entrepreneur Ideas Never Get Built

If you’re an entrepreneur, you know the feeling. Ideas keep you up at night. You see problems everywhere and you can picture the solution, but you can’t build it yourself. So you explain it to someone who can code, and they roll their eyes because they’ve heard enough of your ideas.

For the past two years, I was solving business problems with automations. Zapier, n8n, duct-taping workflows together. I shipped three real products with the help of my developer friends Mahdi and Xana. One of them was a mining hours logging app for a client. Real products, real users.

But every idea had to go through someone else’s hands. And most ideas just stayed in my head.

What Happened When I Tried Vibe Coding

When the vibe coding wave hit, I jumped in. v0, Bolt, Lovable. All the hyped tools. Ironically, I don’t even remember most of their names anymore.

But here’s what they gave me. I learned how to deploy on Vercel. I learned Supabase. I built a lot of apps. I got familiar with the stack.

The problem? I could never get past 90% done. There was always something broken. Always something not working. Close enough to see the finish line, never close enough to cross it.

How Google AI Studio Got Me to a Finished MVP

Google AI Studio was the first tool that let me build MVPs that were actually complete and working. That was a real shift. I was able to build internal tools for myself and even replaced some of my n8n automations.

But I hit a wall. I had no deep technical knowledge, so when I wanted to push an app further, add features, scale it, or fix edge cases, I’d get stuck in painful loops. The AI would go in circles. I’d lose hours.

The MVP was done, but the vision was stuck.

How Claude Code Changed Everything

I was scrolling X one day and saw a post about Claude Code. The idea was simple: what if I export the code from Google AI Studio and keep building in Claude Code’s CLI?

So I tried it. And it failed.

It failed because you can’t build on a bad foundation. The code I exported wasn’t structured well enough to work with.

But the core concept was right. Working with files locally on my system, not in the cloud. Using a CLI. Having actual control over my code.

So I went back to Google AI Studio and asked it to write me a full, comprehensive document explaining everything about the product. The architecture, the database, the features, all of it. Then I took that document and used it to rebuild from scratch with Claude Code.

It worked.

Supabase migrations were easier to fix. Pushing to Git made sense. Deploying to Vercel was smooth. Everything just clicked.

What It Feels Like to Build Without Limits

For the first time, I felt powerful.

If I had an idea today, I could build it today. No more explaining to developer friends. No more waiting. No more hearing “that’s cool but I’m busy.”

The last time I felt this way was in 2020 when I was starting Keyweemotion. That was the pivotal moment in my entrepreneurial journey, and this felt just as big.

I upgraded to Claude Max immediately. Claude Pro wasn’t enough for Claude Code’s usage. And I just started building.

I haven’t stayed up late feeling excited about work in years. This month brought that back.

What I Built: 4 Apps in 30 Days

1. QT. A trading app I built for my client Colin. Now he can share a dashboard with his partners to track affiliates, and it’s connected and synced to their own CRM, GoHighLevel. Everything was manual before. Now it just runs.

2. CMA Report Generator. As part of my internship at RE/MAX, I spotted an issue: agents were spending hours putting together Comparative Market Analysis reports. So I started building a tool where they either upload an MLS PDF or fill out a form and get a polished report in minutes. It’s not done yet, but it’s pretty cool.

3. MoodMarket. My personal passion project. A mood journaling app I’ve been thinking about for a year and a half. I finally built it. It’s real now.

Bonus: OpenClaw. An AI agent deployment I set up to explore what’s possible with autonomous AI systems. I wrote about the full process in my post on deploying OpenClaw for a real client.

Every single one of these started as a conversation. “I wish I had a tool that does this.” “Wouldn’t it be nice if we could track that?” This month, I stopped wishing and started building.

What I’m Not Going to Pretend

I’m not going to tell you everything is perfect.

I’m not confident enough to make most of these apps fully public yet. The casual ones, sure. But the apps handling real business data, trading records, affiliate financials, client information, that’s different. Security is a real concern. If data gets leaked, that’s a real problem for real people.

So that’s my next learning curve. How to properly secure an application. Auth, data handling, encryption, the fundamentals. I solved the building problem. Now I need to solve the trust problem.

What This Means for Non-Technical Entrepreneurs

Here’s the part that keeps me up at night now. A different kind of staying up at night.

I’m not special. I don’t have a computer science degree. I didn’t grind LeetCode. I spent years duck-taping workflows together with Zapier and n8n. But now I can work with CRMs, APIs, and databases to build functioning multi-tenant apps. That’s the transition. From automating someone else’s tools to building my own.

The only thing that changed was the tool.

That means there are thousands of people right now sitting on ideas they can’t build. Entrepreneurs who see problems in their industry every day but can’t afford a dev team. Small business owners who know exactly what they need but have no way to create it.

The wall between “idea” and “product” is crumbling. And if you’re reading this thinking “that could be me,” yeah, it could be.

What’s Next

I’m focused on two things now. Getting these apps production-ready with proper security. And building more solutions for businesses through ConnectMyTech.

The pitch is simple. I understand your business problem, and I can build you a solution. Fast. Not in months. Not with a $50K budget. Fast.

If that sounds like something you need, reach out. I’m just getting started.

Shahab Papoon is an AI & Automation Integrator based in Victoria, BC. Learn more at shahabpapoon.com.