I originally posted this post on my blog a long time ago in a galaxy far, far away.
If you're new to coding, you shouldn't rely on AI to generate code.
AI is here to stay. Sure. We can't ignore it. We have to adapt, like we coders have always done. Absolutely.
But the problem with AI is when we use it to outsource or replace our thinking.
If you're not the one using the tool, you're becoming the tool. And tools are easy to replace when a faster, better, and cheaper tool appears.
If you're a new coder, don't ask AI to generate code for you.
Ask AI to do any of these tasks instead:
- Review your code
- Dissect a piece of code
- Explain difficult concepts
- Create roadmaps and study guides
- Act as your debugging rubber duck
- Generate test cases or sample inputs
- Look for security and performance issues
- Create questionnaires to evaluate yourself
- Learn the most common language features
- Check your understanding of a difficult concept
- Translate one piece of code to another language
- Learn the most common methods from standard libraries
Using AI is like using calculators in math classes. They can make you faster if you know what you're doing, but they can't think for you.
As I learned from Jim Kwik, the brain coach, "use Artificial Intelligence to extend your Human Intelligence, not to replace it."
AI can generate code in seconds. But the real value of a coder isn't in typing. It's in understanding business problems and collaborating to build the right solutions.
That's why I wrote Street-Smart Coding: 30 Ways to Get Better at Coding, a practical guide to the skills that actually make you a confident coder. It's the roadmap I wish I had when I was starting out.
Top comments (2)
Good point. I hadn't thought of using AI that way.
Yes, asking it to generate code is just one task, right?