I originally posted this post on my blog.

The other day, Leon Martin wrote that programming is becoming prompting. He did it here,

And that’s true. The barrier to entry for coding is getting lower and lower.

A few years ago, it was the "Learn to Code" movement. Even Barack Obama replied to a coding question in an interview. One about sorting algorithms.

These days, with AI, anyone is one paragraph away from creating something that works.

Leon made two interesting points in his post:

#1. Too much AI is bad for your health

"...when you start writing prompts instead of functions, you stop flexing those problem-solving muscles that got you into programming in the first place."

Been there! I’ve been using Copilot during my coding sessions. And the other day, I felt the urge to use AI instead of working through a simple problem.

And that’s the real danger of too much AI.

#2. AI needs the right hands

"Knowing how to code is still the superpower. The prompt is just a shortcut."

Absolutely! AI is just like calculators in math class.

You could have the most powerful calculator in your hands during an exam, but if you don’t know what you want to compute, it’s useless. You will fail that exam.

AI is like an Iron Man suit. But it still needs Tony Stark inside it to save the day.

Build the coding skills, then use AI to amplify them. Work on your debugging, problem-solving, and clear communication skills.

If you want a guide to sharpen your coding skills, I wrote Street-Smart Coding: 30 Ways to Get Better at Coding. Some lessons are conventional. Some were learned the hard way. And a few are weird. But all battle-tested.

Get your copy of Street-Smart Coding here. It’s the roadmap I wish I had when I was starting out.