How I Self-Taught Myself To Code (And How You Can Too)

I was first exposed to coding during my early childhood, somewhere around 2008, while playing Roblox.

I never took it seriously until I saw how some people were creating games by blending custom code with a talent for "building" to make something wonderful. I deeply desired to have something of my own. I thought if I could learn this, I could earn tickets to buy any limited item on Roblox.

Well, that didn’t go far. At the time, I didn’t have the discipline to read through Lua documentation and practically apply it. While I learned to build with blocks and ship basic shooter-style games using public templates, it wasn’t good enough. I could get 8-12 people in a server, but they would quickly churn. It was harder than I thought. Eventually, I just gave up because I figured it wasn’t for me.

Fast forward a few years, and I stumbled across code again through necessity.

I was building Shopify stores for my dropshipping business. I wasn't coding from scratch, but I was making minor fixes and trying to understand how pages were laid out. I was curious, but not curious enough to write full pages of flat code. It was more about understanding the structure so I could audit a developer's work or make a quick tweak if I had to. It was about awareness, not mastery.

The Entrepreneurial Shift

During a period of reflection on my skills and what I wanted to do in life, I started exploring hardware (Arduinos, C) and no-code platforms like Bubble.io and FlutterFlow. I actually read their entire documentation just to grasp the basic concepts.

At the same time, I was studying how iconic figures like Steve Jobs and Elon Musk built the future. Most of them were in tech, so I figured a startup was the way to go. I had read Peter Thiel’s "Zero To One" a few years prior without realizing its genius. Looking back, the concepts in that book are critical for any entrepreneur, whether you're in SaaS, e-commerce, or content creation.

Inspired by Y Combinator and the startup world, I came up with an idea in 2022. It was a combination of Obsidian.md and a news aggregator to spot trends which emerge through nodes and network graphs.

I’ll be honest: I made a fundamental mistake. I created a SISP: a Solution In Search of a Problem.

And of course, I got rejected. I sort of gave up on the path again. I felt it would take too long to learn the technical skills required. Back then, if I couldn't figure something out in a relatively short amount of time to ship value, I moved on.

The AI Revolution

Then came late 2023 and 2024. AI became something I took a deep interest in.

For the first time, I realized we had entered a period where you don't actually need to code line-by-line. You just need to direct intelligence. The only limitation became creativity and resourcefulness.

I started doing little side projects, not to monetize, but to experiment. Hooking up APIs, building simple web apps, and seeing what I could do. I grounded myself by reading stories on Starter Story and following indie hackers who were "making it" with no-code or low-code. It proved it was possible.

Then, in late spring 2024, I found Cursor.

This was the paradigm shift. I saw how software could now be created by the "average man." You weren't limited to constrained visual platforms anymore. You could prompt with natural language, and an AI-integrated IDE would code the software for you.

As of late December 2025, the game has changed even further. You can spin up agents to create something for you while you do something else. Come back in a few minutes, and boom, you have a full web app ready to deploy.

The "Orchestrator" Mindset

I think a key mindset that helped me build early businesses was refusing to label myself as a "Software Engineer."

If you label yourself an engineer, you are constrained by that vocabulary. Instead, look at yourself as an opportunist or a missionary.

• An engineer thinks about code.
• An entrepreneur thinks about solving problems, bringing value to society, and buying their own freedom.

I am not technical in the traditional sense. I am an orchestrator. I view AI as a super-powered compiler that transcribes my thoughts and ideas from natural language into executable code.

Steve Jobs reportedly viewed himself this way. He was a conductor. The team was the orchestra, and he coordinated how the music was played. Today, AI is your orchestra. If you know how to play the notes correctly, you can direct it to play any piece of music you want.

How to Learn: Throw Yourself into the Fire

The best way to learn, and I still strongly believe this, is to throw yourself into the fire and build.

Force yourself to create a project that motivates you. If the goal is financial freedom or a successful exit, you will find solutions to the deficiencies in your learning. You will be forced to solve every problem you run into.

Yes, it is overwhelming at first. Opening a code editor, figuring out deployment, understanding frontend vs. backend. But if you solve a specific set of problems enough times, that data becomes "chunked" in your knowledge tree. You develop an intuitive grasp of how things work.

My football coach used to always say: "Figure it out."

That trait has served me better than any technical class could. The more projects you ship, the more your brain rewires itself. After nearly two years of coding with AI, I don't ask if I can code something. I have a gut feeling. I look at my skills, my resources, and the tech stack, and I know if it's within my capacity.

The Synthesizer

To win in this new era, you need a synthesizer mindset.

Don't stick in your lane. Hang out on X, join Discord communities, and see what people are building. Treat the world as a playground of resources. When you have a stream of high-quality information, you can piece things together (skills, APIs, ideas, people) in unique ways that haven't been done before. That is innovation.

Just Do It

Software is now democratized. A teenager, an elderly parent, or a complete civilian can prompt AI and create the future.

Don't look at this as something scary. Look at it as being the orchestrator of a superintelligence. It doesn't mean you shouldn't learn the basics. Understanding what a database is or how a frontend talks to a backend allows you to direct the AI more precisely. But you no longer have to do the heavy lifting.

One of the most powerful things you can do in the 21st century is learn to code with AI. It just takes one idea, one shipped project, to change your life.

Use AI to learn. Build the future. Figure it out.