I’ve been using Synara for a while now, mostly across my portfolio and a few things I’m building.
And I noticed something.
I’m spending less time telling AI how to code something and more time telling it what feels wrong.
A small example. I noticed a button in my portfolio was missing its hover state.
Instead of finding the component, checking how other buttons behave and fixing it myself, I just said:
Nothing groundbreaking happened here.
But this small interaction made me realise how much my workflow has changed.
Before, AI coding still felt like pair programming through a chat window.
Prompt.
Generate.
Check.
Fix.
Prompt again.
Now I’m slowly moving away from that.
I give Synara the problem and the context. It figures out where to look, what already exists and what needs to change.
I spend more time looking at the decision and the diff than thinking about the implementation.
As a designer, this is probably the part I find most interesting. I can stay closer to the product.
Something feels off → describe it → let the agent investigate → review what changed → move on.
The code is still important.
I’m just spending less of my attention on writing it. And more on deciding what should exist in the first place. That feels like a much more interesting way to build.
Understand deeply. Design only what matters.