I’m not a software engineer.
Honestly, with all the new tools, model releases, and tech jargon, AI can feel overwhelming. Part of what I want to do here is make it feel less intimidating by testing these tools in real life, figuring out what is actually useful, and sharing what I learn along the way.
For me, technology is also a way into something bigger. I’m interested in how new ideas help us explore the world, learn from people in different places, and connect things that usually sit in separate boxes. If AI makes it easier to follow our curiosity and turn those connections into something tangible, I want to understand how.
But I’m building interactive websites now.
One recent experiment combines moving words, physics, sound, and imagery inspired by rooftops around the world. The whole thing takes less than two hours to get working.
The idea starts when I see Marina, an UX design influencer, experimenting with hanging “curtain words.” Her version uses Framer and a few different steps. I love the interaction, but what catches my attention is everything around it.
What happens when the words move more like wind chimes? What happens when the experience takes you through rooftops inspired by different parts of the world? What happens when a website feels less like a page and more like a place you explore?
Instead of recreating the same workflow, I give the idea to Qoder, the Chinese AI coding tool I’ve been experimenting with.
Qoder handles most of the imagery, web design, and code. I spend more of my time shaping the experience: changing visuals, fixing behavior that feels off, and adjusting the movement and sound until it gets closer to what I have in my head. Three things stand out.
1. You start with the experience, not the technical solution
This is probably the biggest change for me. I don’t begin by choosing a language, framework, or technical architecture. I begin with a feeling:
I want this to feel like walking through an interactive collection of places around the world.
Then I describe what I want to see, how I want it to move, and how someone should experience it. Qoder works backward from there. I give it the task and context, and it handles much of the implementation.
That changes the starting point. The question is no longer, “Do I know how to build this?” It becomes, “What do I want this to feel like?”
For anyone with a folder full of half-formed ideas, that is a meaningful shift. You can start with the thing you want to make rather than the technical steps standing between you and it.
2. AI does the heavy lifting, but taste still lives in the details
The first version worked. But it also isn’t quite right. The hanging words were one of the places where I spend the most time editing. The interaction uses pendulum-style, constraint-based physics. The words hang from fixed points and swing when they’re disturbed, which gives them the feeling of physical chimes rather than static text.
Making something move is relatively easy. Making the movement feel right takes more judgment.
The same applies to the sound, imagery, spacing, and atmosphere. AI gets me to a functioning version quickly, but I still decide what feels generic, what feels awkward, what needs more movement, and what needs to disappear.
That’s why I don’t think the most interesting AI skill is prompting. It’s judgment.
You need enough taste to know when something technically works but still feels wrong. You need enough reference points to recognize when the output looks like everything else. And you need enough confidence in the original idea to keep pushing instead of accepting the first version. The tool speeds up execution. It doesn’t make those decisions for you.
3. Curiosity and taste are becoming practical, prized skills
This is the part that interests me most.
The project doesn’t start with, “I should build a website.” It starts with questions. How do words feel physical? How does a website capture some of the feeling of discovering another place? What changes when sound becomes part of the interaction? What happens when someone without a software engineering background starts building anyway?
Less than two hours later, I’m clicking through something I can actually see, hear, and interact with.
Which is key for now and the future- Technical knowledge still matters. But curiosity and taste are becoming more valuable alongside it. Curiosity helps you notice possibilities other people overlook. Taste helps you decide which of those possibilities are worth pursuing and whether the result is actually good.
History, design, languages, culture, science, business, art, and technology stop sitting in separate boxes. You start combining them into things other people can experience. A question becomes a prototype. A visual reference becomes an interface. A passing observation becomes something tangible. And that’s a very different way to think about learning AI.
Instead of trying to memorize every new tool, you get better at noticing what interests you, forming better questions, and making something from them.
What are you building right now?
Add to Your Queue!
AI & Technology
On My Radar
On My Radar Dispatch: August 10, 2026 - The US-China Tech Squeeze, Silicon Valley Lawsuits, Hyperscaler Agent Hacks, Rumors, Implementing AI in your Workflows and the Jobs Coming Next
Thanks for reading Potentia! Join 3.5k+ followers!
We genuinely appreciate every like save and follow : )
Your support means the world!
Connect with us on Youtube, Twitter/X , and Instagram, Apple, Spotify!