Lately I’ve been using AI a lot while coding.
But not in the way most people think.
I’m not opening AI and saying “build this for me”.
Most of the time it looks more like this:
Real Scenario 1 — Getting Stuck
I’m writing something simple, like a layout or a component.
At some point, things stop making sense.
Instead of dumping everything to AI, I usually write something like:
“I’m trying to do X, but this part feels wrong. Can you point out what I’m missing?”
Not asking for code.
Just asking where I’m thinking wrong.
Most of the time, that’s enough.
Real Scenario 2 — Something Works, But Feels Wrong
Sometimes the code works.
But I know it’s not clean.
That’s when I do this:
“This works, but I feel like it’s messy. Where is the problem?”
And AI usually points out things like:
- repetition
- bad structure
- unnecessary complexity
That’s useful.
Because I already solved the problem — now I’m improving it.
Real Scenario 3 — I Don’t Understand My Own Code
This happens more than I’d like to admit.
I write something, it works…
but if you ask me to explain it, I’d struggle.
So I ask:
“Explain this like I wrote it but didn’t fully understand it.”
That’s where AI becomes actually valuable.
What I Don’t Do
I don’t do this anymore:
- “Build me a full app”
- “Write everything from scratch”
- Blind copy-paste
Because I tried that.
It works short term.
But after a while, you realize:
You’re not improving — you’re just producing.
The Only Rule I Follow
Before I ask AI, I ask myself:
“Did I actually try to solve this?”
If the answer is no, I don’t open AI yet.
Small Change, Big Difference
The biggest change for me was this:
Instead of asking AI to do things for me
I started using it to question my thinking
That’s it.
Final Thought
AI is powerful.
But if you use it the wrong way, it just makes you faster — not better.
Right now I’m trying to use it in a way that:
- makes me think more
- not less
Still figuring it out.
Top comments (0)