How to Learn to Code in the Age of AI

When I first learned to code, I was 23. No one in my family programmed. I had no degree. And the most powerful tool I had was Google.

That's changed — and honestly, it's changed everything.

The myth I want to kill first: coding is not reserved for a certain type of person. Not for men, not for people who've been at a computer since they were eight, not for anyone with a specific background. Coding is for everyone. Including you.

But here's what I'd tell you today that I couldn't have told you five years ago: you don't have to figure it out alone anymore. AI tools like Claude and GitHub Copilot have fundamentally shifted what it means to learn to code — and if you're starting now, you're actually starting at the best possible time.

---

Step 1: Figure Out Your Why

Before you open a single tutorial, get honest with yourself about why you want to learn.

Do you want to build an app? Automate something repetitive at work? Understand the engineers you manage? Create a website? The answer changes everything — which language you pick, which tools you use, how long it'll take.

When I started, I was building DecorRaid and needed to understand machine learning algorithms. That "why" pointed me straight to Python. There was no point in learning something else first.

Your why is your compass. Don't skip this step.

---

Step 2: Choose a Language That Matches Your Goal

Once you know your goal, the language usually picks itself:

• Python — data, AI, automation, machine learning, scripting
• JavaScript — web apps, interactive websites, frontend and backend
• SQL — databases, analytics, working with data
• Swift / Kotlin — native mobile apps

Don't spend weeks agonising over this. Pick the one that matches your goal and start. You can always learn another later — and the second one is always easier than the first.

---

Step 3: Get the Basics in Your Body

You still need to learn the fundamentals. No shortcut here. Find a beginner resource — a book, a YouTube channel, a short course — and work through it with intention.

Install your tools early and get comfortable in your environment. For Python, something like Jupyter Notebook or VS Code. For JavaScript, open a browser and start. The earlier you write real code (even tiny, broken code), the faster you learn.

Start with a small, motivating project. Not a tutorial exercise — something you actually care about. Even something simple. That personal investment is what keeps you going past the hard days.

---

Step 4: Use AI as Your Coding Partner — Not a Crutch

Here's the big update. When I learned, I Googled everything. And that was genuinely good advice — knowing how to search, how to read Stack Overflow, how to parse documentation — these are still skills worth having.

But today, you also have something much better available to you: Claude and GitHub Copilot.

GitHub Copilot lives inside your code editor. As you type, it suggests completions, fills in functions, and helps you spot patterns. It's like pair programming with someone who has read every library ever written. You still need to understand what it suggests — but it dramatically accelerates you past the "how do I even start this" phase.

Claude is more like a patient senior developer you can talk to. You can paste in broken code and ask "what's wrong here?" You can describe what you're trying to build and ask "what's the best approach?" You can ask it to explain a concept five different ways until one of them lands. It doesn't judge you for asking beginner questions.

A few ways I'd use these tools if I were starting today:

• Stuck on an error? Paste it into Claude and ask for an explanation, not just a fix. Understanding why it broke is the lesson.
• Writing boilerplate? Let Copilot handle it. Your job is the logic, not the ceremony.
• Learning a new concept? Ask Claude to explain it with an example that matches your actual project.
• Code review? Ask Claude to critique your code and suggest improvements. It's more honest than most humans.

The key mindset: use these tools to learn faster, not to avoid learning. If you copy code you don't understand, you're building on sand. If you use AI to help you understand, you're building something real.

---

Step 5: Break It Down and Keep Moving

Every big goal needs to be broken into embarrassingly small steps. Not "build an app" — "make a button appear on the screen." Not "learn Python" — "write a script that prints my name."

Small wins compound. They also tell you what the next step is.

The people who succeed at learning to code aren't the ones who are naturally talented. They're the ones who keep going past the point of confusion — and now, you have better tools to help you through that confusion than any previous generation of learners had.

---

The Bottom Line

Coding is still about problem solving, curiosity, and patience. Those things haven't changed. But the experience of learning has changed enormously.

You no longer have to wade through ten Stack Overflow threads to understand one error message. You don't have to wait to find a mentor. You can have a back-and-forth conversation with an AI that knows more than any single person you could ask.

Start with your why. Pick a language. Learn the basics. And use every tool available to you — including the brilliant AI ones that exist right now.

You've got this.