I Was Quoted $14,000 and 4 Months. I Built It Myself in 4 Days.
Why This Project
When I started this project, my dad was battling cancer. He later underwent a liver transplant, and now he’s still fighting for his life.
He’s spent decades as a business consultant teaching frameworks, strategies, and ways of thinking that shaped hundreds of thousands of leaders, and at some point I realized I didn’t want all that wisdom to just disappear.
A while back, I built an AI chatbot trained on his teachings as a way to preserve his voice and knowledge. I wrote about that initial journey in I Built an AI Version of My Dad to Keep His Legacy Alive.
The web version worked fine, but the people who could really benefit are in China. In China, if you want to reach people, you need to be on WeChat. With over 1.4 billion users, it’s the default platform for living in China: shopping, communication, travel, payments. To reach users effectively, you need a Mini Program — an app that lives inside WeChat.
Starting From Scratch
I thought I had a head start. I’d built MySpotipal before, where I learned my way around APIs, AWS EC2, and full-stack development.
But WeChat Mini Programs run on a completely different stack. The frameworks and deployment flows were created by Chinese tech companies. Suddenly, all the technical terms I’d used in English had to be processed in another language.
The discomfort with new technology came rushing back.
How do I use Chinese cloud technology? How do I navigate Tencent’s developer console? How do I design UI for not even a mobile app but a mini program?
Each question opened three more, until I felt completely overwhelmed.
The Doubt
For a moment, I almost gave up.
My brain spiraled:
What if people actually use this?
What about databases, server capacity, data security?
What if it breaks? What if it’s slow? What if I mess this up?
I didn’t have a team of engineers, or funding, or much experience with Chinese infrastructure. The more I thought about it, the more convinced I became that this was beyond what I could reasonably figure out on my own.
Out of curiosity, I reached out to a few companies for quotes. The most expensive came back at $14,000 and four months.
At first, that number stung. Fourteen thousand dollars felt absurd for something I believed I could learn. But I didn’t have that kind of time or money, so I figured I might as well try.
The Build
The turning point came when I dug more into Tencent CloudBase, their serverless cloud platform. It handled the databases, hosting, and deployment — essentially all the infrastructure work that had been keeping me up at night. Suddenly, the project went from “impossible” to “maybe doable.”
So I started what became another solo vibe coding session: four days and nights of laser focus. Reading documentation, breaking things, fixing them, and breaking them again. The AI part was easier than expected thanks to Tencent’s built-in no-code agent tools.
The UI, however, was a struggle. I’d sketch something out, implement it, look at the result, and immediately hate it. Everything I made looked amateur, like someone’s first attempt at a mobile app, which it was. Eventually I swallowed my pride and reached out to some designer friends for feedback, spent time browsing for inspiration online, and found Figma Make with AI assistance, which helped me get to designs that were at least okay-ish. Not beautiful, but functional enough that I wasn’t too embarrassed by them.
I kept catching myself obsessing over details that didn’t matter yet — the exact shade of a button, the spacing between elements, micro-alignment nobody would notice. Each time, I had to pull myself back to the real goal: just get the beta out.
Perfection is the enemy of good.
After four days, the Beta version was live. I opened WeChat and there it was…running on the same app I use every day.
When my dad first tried it, he was more enthusiastic than anyone else. He spent hours exploring prompts, and smiling at the responses. “The answers are good,” he said. “I like that it pulls from the internet.”
In one of Ali Abdaal’s videos, one quote stuck with me for a while:
If there is something that someone has done, given the time, effort and work, I could replicate at least a little bit of that success.
When I first heard this, I did not believe it to the very least.
But this time, I did. For the first time in my life, I felt a genuine, grounded belief in what I could accomplish if I was willing to put in the time, effort, and work. I’d looked at something that felt way beyond my reach, been told it would take months and thousands of dollars, and built it anyway in four days.
Imperfect, yes. But real, functional, and meaningful.
What I Learned
Building with AI tools today makes rapid prototyping so much easier.
Every project like this rewires your brain. You stop asking, “Can I figure this out?” and start knowing, “I’ve done it once, I can do it again.”
The Mini Program isn’t public yet since we have not figured out a monetization path. But the beta exists and my friends and family can use it.
That ended up meaning more to me than I thought it would. Sometimes what looks like a wall is just a door you haven’t tried to open yet.
