I Built My Daughter a Game in 2 Hours And It Changed How I Think About Software
- Kish Melwani

- 4 days ago
- 3 min read
Last weekend, I did something I never thought I'd do: I built an app from scratch. Not a simple calculator or to-do list — a fully designed, mobile-responsive fan game for my daughter based on her favorite artist, Melanie Martinez.
I did this mainly for research. I'm not a programmer. I'm a business owner. I run an MSP. My days are spent solving technology problems for other people, not writing code. But I sat down on a Saturday afternoon, fired up an AI coding tool called Claude Code, and two hours later my daughter was playing a game I made just for her.
The look on her face made the whole thing worth it.
What the Game Actually Does
The app is called "Melanie's Dollhouse," and it's a music trivia game built around Melanie Martinez's discography — her albums Cry Baby, K-12, and PORTALS.

Here's how it works: a song clip plays, and you have to guess the title from three multiple-choice options. You pick a difficulty level that determines how much time you get to answer. If you get several right in a row, you earn bonus streak points. Each game runs 10 rounds, and your score lands on a leaderboard so you can compete with friends and family.

Beyond the game itself, there's an About section with a timeline of Melanie's career and a media gallery where you can browse her discography and watch music videos organized by album.

The whole thing is mobile responsive, so my daughter grabbed her iPhone and started playing immediately. That was the real test — and it passed.
The Process: "Vibe Coding" Is Real
I'd been hearing about "vibe coding" — the idea that you can describe what you want to an AI and it builds it for you, then you just refine and iterate until it's right. I was skeptical. But I tried it, and honestly? That's almost exactly how it went.
Most of the work wasn't writing code. It was looking at what was being built, deciding what needed to change, and describing those changes. Make the colors more playful. Add a streak counter. Make it work on mobile. Add a leaderboard. It felt more like directing a project than engineering one.
The tool I used was Claude Code, which runs at $20/month. For context, that's less than most of the SaaS tools collecting dust in your tech stack.
Why This Matters Beyond a Weekend Project
Here's where it gets interesting — and honestly, a little unsettling.
If someone with no programming background can build a polished, functional, mobile-responsive web application in a single afternoon, what does that mean for the software industry?
We've already seen the impact. Software stocks have taken hits. The conversation about AI replacing developers is no longer hypothetical — it's playing out in real time. I'm not saying developers aren't needed. Complex systems, enterprise software, security-critical applications — those still require deep expertise. But for a huge category of business tools, internal apps, customer-facing utilities, and workflow automations? The barrier to entry just collapsed.
For years, small businesses like mine have adapted to software. We buy a platform, and then we reshape our workflows to fit the way that platform thinks we should operate. We pay monthly fees for features we don't use, deal with interfaces that don't match how we actually work, and accept limitations because "that's just how the software works."
That era is ending. Now the software can adapt to us.
What This Means for ServiceByte
I'm planning to take what I learned and apply it to our internal operations. Custom tools that fit exactly how our team works. Client-facing utilities that solve specific problems without the overhead of a full SaaS subscription. Internal dashboards that show us exactly what we need to see, nothing more.
This isn't about replacing our existing tools overnight. It's about filling the gaps — the small, specific problems that no off-the-shelf product quite solves, the workflows that are "good enough" but could be great.
The Takeaway
It's fascinating and a little scary to see this much change happening this fast. But if there's one thing I took away from a Saturday afternoon building a game for my daughter, it's this:
If you can dream it, you can build it. And if you can build anything you dream of, why would you keep following someone else's vision for how your business should run?
I'll be sharing more about this journey as we experiment with AI-powered tools at ServiceByte. If you're a business owner curious about what's possible, feel free to reach out. I'm happy to talk about the experience.


Comments