The Era of Machines Begins...

The big moment has arrived for the GT fans (the famous green texts): the era of machines begins!

Jokes aside, I wanted to start this blog by doing what I do best: being a clown.

With that said, in this post, I want to talk a bit about AI (shocker, right?), but from my point of view and based on what I’ve been experiencing.

How was this blog born?

I can start by saying how AI has made it easier for people with little to no knowledge to create new things. This blog right here was built using the Hugo Framework, which I was honestly too lazy to read the documentation for. The idea came from a desire to build a personal brand, and I thought having a blog to share knowledge and reflections would be a good place to start. I wanted a more “programmatic” design without losing the minimalism, so I took inspiration from this resume style:

The C# fans out there probably just hit the roof. Anyway, since I’m on Team Kotlin, I wanted something more tailored to that. So, I ran some prompts to create this design on Super Design, which spat this out:

I thought it looked promising. From there, I installed Hugo on my machine and practically vibe-coded the whole thing using GPT-5.3-Codex through a VS Code extension. I made a few fine-adjustments and some structural improvements, but I’d say 98% was done by AI with ridiculously simple prompts. My only knowledge of Hugo up to this point is how to create posts, publish them, and deploy a new build (other than that, I already had some experience with handlebars/jinja).

This story of how I created this blog serves to show us software engineers how AI democratizes programming, graphic design, and much more.

What do I think about all this?

There’s no point in being a denier; I believe we’ll reach a point where we’ll almost never need to write a line of code again. Models are constantly improving their code output for the most popular languages and frameworks. I think this might “separate the wheat from the chaff”—devs who rely solely on knowing how to code will likely suffer with every new LLM advancement.

“Ok, so how are you adapting to all of this?”

Well, I’ve always been a developer who tried to understand the “whys,” and I feel that this is becoming increasingly MANDATORY.

That being said, I try to understand at least the basics of how an LLM works, how to write good prompts, how to create agents, and everything else that surrounds my work as a software engineer: requirements engineering, how to build testable and maintainable software, design patterns, best practices, scalability, and load testing… the list is long.

But take it easy—you don’t need to learn everything at once. Go at your own pace. Learn what your job requires of you, see how you can be more productive using AI, test things out, make mistakes, and improve.

Need to master a framework to perform better? Use AI to explain its core concepts, ask it to list the most important features, or have it create challenges for you to learn how it works. And even if your framework isn’t that well-known, there’s RAG and other alternatives to give the AI the context it needs. Use it to learn and to perform.

Now, the million-dollar question: was this post made by an AI?

Negative. The entire creative process came straight out of my noggin to my typing fingers on the keyboard. No generic content around here, my friend ;)