Prompting NEVER Make You Learn Faster

As I stated in my previous post, PROGRAMMING IS VERY HARD. And suddenly from the sky falls a magical genie promising to flatten this difficult path - who wouldn't be tempted? ChatGPT, Gemini, Claude and 1010 LLM Wrappers from heaven appear playing the role of genie... the most powerful saviors in history, always ready to provide answers in seconds.

The omnipotent genie can answer everything under the sun and beyond. Even for things that don't exist, problems that aren't real, the genie is ready to stroke your ego and worship the wisdom of the questioner.

YOU'RE JUST WATCHING SOMEONE ELSE WORK OUT

Think about it seriously. When someone wants to develop muscle, they have to push their workout to the limit, to the threshold where every muscle fiber tears. It's precisely during the process of the body healing those training tears that muscles become bigger, thicker, and stronger.

PROGRAMMING THINKING IS EXACTLY THE SAME

Those tears are the moments when you face program errors, the hours of deep immersion to optimize a problem, the feeling of being stuck when the logic you wrote runs contrary to expectations. Each time you debug an error yourself, find a better solution yourself, is a time when neural connections in your brain create a new pathway, a deeper, more solid groove in your thinking.

This muscle-building process is painful, slow, but irreplaceable.

The AI genie, with its infinite kindness, has robbed you of the process of creating those valuable scars.

You won't learn anything from perfection that falls from heaven.

BUT AI MAKES MISTAKES TOO. I CAN LEARN FROM AI'S ERRORS

The problem is that to recognize mistakes somewhere, or from someone, you need a certain foundation of knowledge.

When AI writes wrong code, here's the difference:

✅ Senior: recognizes immediately, chuckles and can share it with colleagues as a funny professional story.

❌ Beginner: completely lost. They don't know if they're wrong or AI is wrong. Next they ask another AI to check, or worse, type a new prompt and pray for luck.

Suppose after n prompts, AI finally got it right. Are you sure this is a correct result, and the working method to reach this result is quality?

You don't have a good compass, and you're lost in the AI fog but think you're demonstrating navigation skills 4.0. Complete waste of time and money.

BUT AI IS STILL AI, WHAT SHOULD I DO?

More to come.