I want to be honest about something.
I started building this website with a clear goal — a site that loads in less than 1 second, and would still give me a decent content management system to make publishing fun and easy.
According to my AI assistant, building this blog was supposed to take an afternoon. It took longer. Not because the technology is hard — it isn't. But because it kept making things sound simpler than they were.
The Pattern
Every time I asked a question, the answer was "easy." Every feature I wanted was "totally doable." Every concern I raised was met with confidence.
And then reality hit.
Commands that didn't work. Files in the wrong folders. Three different locations for the same project. Terminology I didn't understand. Promises about features that turned out to be entire separate projects.
Trust me, the gap wasn't in "tech terms" — it was in communication models. Creating and bringing the concept of this site to life was a "right brain" activity while my AI assistant kept sending "left brain" messages, until I stopped it.
What Actually Frustrated Me
Not the terminal. Not GitHub. Not Vercel.
The gap between what was promised and what was delivered. That gap is exhausting. It makes you feel stupid when the problem was never you — it was the expectations being set.
To be completely honest, the first build failed miserably. I'm not even talking about the $150 of failed experiment cost paid to the vibe coding app. I'm talking about time waste, energy, and an entire weekend gone.
Why I Kept Going
Because underneath all the noise, the thing actually works.
The site is fast. It looks mostly like how I wanted. The design is mine. And now that it's running, the workflow is simple enough that I'll actually use it.
AI, through its widespread applications, is no different than other business and life skills. You can't expect to try something once and master it out of the blue. The key is to try again and find how to make AI work for you — and how to co-create with it.
What I'd Tell Anyone Starting Out
Don't let the AI tell you something is easy. Ask it to show you exactly what "easy" looks like before you agree to anything. Ask for mockups, visuals.
And when it frustrates you — because it will — remember that the frustration means you're building something real.
There are so many posts online celebrating results and "breakthroughs" with AI, but so few mention the hours, challenges, and failures involved in getting there. So I just want to put it out there: if your results are mediocre, if you feel overwhelmed and struggling, just try again. Simplify the scope, experiment with something smaller, simplify the prompt, remove any expectation of "usable" results, and treat the whole thing as a fun experiment.
That's worth it.