RightTech

The small build habit that saves me hours is boring...

I fix one thing at a time.

When I’m working with AI on a bug fix, the worst move is throwing five problems into one prompt and hoping for a clean result.

That’s how a quick repair turns into a full afternoon of chaos.

AI coding tools like Claude Code and Codeex are good enough now that literally anyone can now build software.

We’re entering an age where it’s getting even easier for people.

But easier to build does not mean easier to debug if you get sloppy.

Here’s the simple way I handle it:

1. Pick the single problem that hurts most.
If the login is broken, start there.
Not login, layout, emails, and reporting all at once.

2. Describe what is happening in plain English.
What did you expect?
What actually happened?
What error showed up?
That gives the AI something clear to work with.

3. Ask for one fix.
Not a redesign.
Not “clean up the whole app.”
Just one issue.

4. Test that change before moving on.
If it works, great.
Then do the next problem.

This habit matters because the bottleneck is no longer skill.

It really is imagination... and discipline.

You can basically do anything you can imagine with these tools.

But if you stack too many requests together, you make it harder to see what actually fixed the problem... or what broke something new.

Simple builds stay simple when your fixes stay small.

How do you handle debugging with AI right now?

The shift hit me all at once...

Software stopped feeling like something reserved for technical people.

It started feeling like something I could actually build.

For years, I treated software like a specialist sport.

You needed a dev team.

A budget.

Time.

A lot of patience.

Now... literally anyone can build software.

That sounds like hype until you try it.

What changed for me was seeing tools like Claude Code and Codex take a plain English idea and turn it into something real.

Not perfect.

Not some giant SaaS platform built by a team of 50.

But real enough to test.

Real enough to use.

Real enough to help a business grow.

That was the mental shift.

Software and the cost of software should not be the limiting factor anymore.

We’re entering an age where it’s getting easier fast.

The bottleneck is no longer skill.

It’s imagination.

And the willingness to test a small idea instead of just talking about it.

That’s why Damian and I got so excited about this.

So many people had a barrier to entry before.

Now that barrier is much lower.

You can build with tech, not teams.

You can start with one messy problem.

One workflow.

One annoying task.

Then see if AI can help you turn it into a simple app.

That’s the moment things changed for me.

Not when I became technical.

When I realised I didn’t need to be.

What small thing would you build if you stopped assuming software was out of reach?