You know by now that AI can dramatically speed up your development process (when used correctly.)

But the key is knowing how to communicate with the AI properly.

Here’s my collection of prompts that actually work in real-world scenarios.

The first time I encountered Big Tech was at age 15 when I won Google Code In. They flew me and my family to San Francisco and showed us around the Googleplex. I arrived with wide eyes, eager to see where the “smartest people in the world” worked.

But, what I found… disturbed me.

Everyone wore the same badges, slept in nap pods, played the same games, and ate at the same cafeterias. I couldn’t escape the realization that I was looking at a daycare for adults.

That day, I silently promised myself I would never work in such an environment.

Imagine if you could build your own apps and tools just by describing what you want. That’s now possible thanks to new AI tools.

This guide will show you how to do it! This guide is specifically for beginners who have never done anything technical before.

What We’ll Cover

1. How to turn your ideas into real working apps
2. Which free tools to use
3. Step-by-step instructions for your first project
4. Common mistakes to avoid
5. How to make your creation available to others

Is it just me, or are the code generation LLMs we’re all using not that good?

For months, I’ve watched developers praise LLMs while silently cleaning up their messes, afraid to admit how much babysitting they actually need.

I realized that LLMs don’t actually understand codebases — they’re just sophisticated autocomplete tools (with good marketing.)

After two years of frustration watching my AI assistants constantly “forget” where files were located, create duplicates, and use completely incorrect patterns, I finally built what the big AI companies couldn’t — or wouldn’t.

I decided to find out: What if I could make AI actually understand how my codebase works?

AI is incredibly powerful, but it needs guidelines. “Vibe coding” might work initially, but as the project grows, it creates more mistakes than it solves. After fixing countless AI implementations, I’ve distilled it down to three core principles that actually work.

The current wave of AI tools promises to 10x your development speed. What they don’t mention is how they can also 10x your debugging time if implemented poorly. I’m building tools to solve exactly this problem, and I’m sharing some lessons I’ve learned along the way.

I recently moved to NYC since I was accepted by the Recurse Center, and today was my first day at their hub.

The day started by nerding out on the retro computers, hardware labs, and 3d printers they have; followed by the first breakfast bagel of my life. I feel incredibly fortunate to be surrounded out by so many talented programmers and am looking forward to the next three months with my new friends!

Last Tuesday at 1 AM, I was debugging a critical production issue in my AI dev tool. As I dug through layers of functions, I suddenly realized — unlike the new generation of developers, I was grateful that I could actually understand my codebase. That’s when I started thinking more about Karpathy’s recent statements on vibe coding.

For those who missed it, Andrej Karpathy recently shared his thoughts on what he calls “vibe coding” — essentially surrendering code comprehension to AI tools and hoping for the best. His exact words? “I ‘Accept All’ always, I don’t read the diffs anymore.”

I have learnt a lot from Karpathy and use AI tools daily, but there’s a world of difference between augmenting your capabilities and completely surrendering your understanding.