Learning How to Code in the AI Age

A full guide for beginners to learn how to code in the AI era

Is learning how to code still worth it?

Many beginners are hesitant to learn to code due to the evolution of generative AI. Watching an advanced chatbot produce a long working code in almost any programming language causes a huge motivation drop to start any programming tutorial, especially among beginners. The good and surprising news is the best time ever to learn how to code is the AI age. I will explain to you why:

AI kills human skill

Let's explain this like a math formula: More people depend on AI to generate code = fewer people use their brains to generate code.

The best way to learn programming skills is to use your own brain. When people prompt an AI to generate a simple website in less than 1 minute, their brains work in the passive mode, which means they don't learn anything unless the AI is used for learning, which is another story. This habit is accumulating over time and kills human skill over time. Let's explain it more in a scientific POV: The best learning technique to learn anything, including coding skills, is the Active Recall technique, which means simply retrieving information from your brain to move what you learned to the long-term memory. When you teach something to people, write something, or remember something. You are actually learning. Using AI to generate code instead of you, without understanding the code, won't help you learn; it wastes your time in your programming learning journey and reduces your overall productivity. Repeating the habit will destroy your learning career in the long term.

Why do we learn programming if AI can do the job?

You may ask yourself this question after reading the last paragraph: If AI is smart enough to produce a working code, why do we learn how to code? Humans invented machines to help us. Imagine a math class without a calculator. Learning how to add, subtract, multiply, and divide big numbers or even small numbers is a hard skill. This is why the calculator was invented: to make our lives easier. But calculators, even the programmable ones, won't replace a mathematician and do their job. The mathematician's job is more complex than just doing basic or advanced math operations. Thinking that AI will eliminate programmers is like thinking that calculators will eliminate mathematicians. The job of a programmer is not just producing working code. The human programmer can produce clean code, which even an advanced AI can't do perfectly at the current time. Humans have the ability to follow a company's style and to see engineering bugs that AI can't see. AI is still developing; it will be great if AI can match human skill, but we can't predict the future.

You may feel productive when a generative AI outputs several pages of code in less than a day. WOW. But you may be disappointed later if you find a bug that can't be fixed by AI, which will waste more time than the time saved initially. You may give up and cancel the whole project. I faced the same situation when I was working as a freelancer. Imagine a whole month of AI coding and another two months just to fix a bug, but unfortunately, I couldn't, so I canceled the project. The client asked for a refund. I felt extremely disappointed. I have never faced this situation before. In the past, when I coded my projects manually. My brain learned and indexed the codebase. When I found a bug, I fixed it quickly, because I know how my code works, so it was easy to debug the code and fix the bug. When I started using the AI, I pasted a code from a chatbot I didn't understand. If I found a bug in a large codebase generated by AI. Only AI should fix the bug. because the codebase is on the AI database, not in my brain. If AI couldn't fix it, I couldn't fix it either. This situation in a critical project causes a lot of stress. There are two solutions here; each one is time-consuming: 1-Understanding the AI-generated codebase. 2-Rewriting the whole project manually. The first solution is a pain, especially among programmers who hate reading other programmers' codebases. Reading clean code is a must-have skill because you are supposed to work within a team, but many programmers prefer the solo path to avoid reading spaghetti code or for other reasons. Some AI-generated codebases are worse. The second solution is the healthy one, but it costs time. At this point, you wish you had started the project without relying on AI.

Programmers are still in high demand

Do current companies hire human programmers or AI agents? We are living in the real world, not the Matrix movie!. AI technology is developing, but don't be scared. It's unlikely in the near future that companies will post a job offer to hire only AI agents. Humans have emotions and social skills; AIs are emotionally stupid, with almost zero social skills. AI doesn't care about deadlines; it can destroy the entire company's value by producing a stupid bug and say: Sorry to hear that. How can I help you today?

So in the AI age, creativity and social skills have important value as logical skills. This is what differentiates you from AI, and this is why companies are still hiring human programmers and giving them high salaries.

How to learn how to code?

Learning is a general human skill; it's not specific to coding only. If you understand how this skill works, you can learn anything, including how to code.

Scientifically speaking, how to learn anything?

Most people hate science. I will say one magic word without wasting your time with fluff: it's easy to remember; it's called Active Recall. I have mentioned it earlier in this post; let's recall it! Can you remember what I said about it without looking at the post?

This is a short practical lesson about active recall, which is simply using your brain to retrieve information without looking at the information source. I will give you just a few examples to understand more.

• Teaching something to someone helps you understand more.
• Writing about a topic helps you learn it better.
• Using a physical or digital flashcard tool like Anki helps you learn efficiently.

The best single way to learn programming

The best way to learn how to code that matches the scientific active recall technique is to do projects without writing a single line of code with AI. Working on projects forces you to explore tutorials and documentation and write the code from your brain. This lets you apply the active recall technique. Just make sure not to generate code with AI except for explanation. Active recall has 2 steps: the brain's input and output. Input happens when you explore tutorials, documentation, or chatbot explanations. Output happens when you write the code yourself without copying and pasting human or AI code. So the correct use of AI is to teach you, not to write the code for you. It's OK to produce code from AI to learn from it, but the final code must be written by you. This is the only way and the best way to learn how to code in the age of AI.

Which programming language to start with?

Your main goal as a beginner is to learn programming concepts, not a programming language. Most programming languages share the same concepts, such as printing information, if condition, for loop, etc., but each language has a specific syntax. The best example to understand this is by our spoken languages. All human languages share the same concepts, but they differ in the way of writing, for example: The sky is the same in Morocco, Spain, the USA, China, etc., but it's written differently in each language. Programming languages are similar.

So learning one language will help you learn others, but as a beginner, you should pick a simply written language to focus on concepts rather than a complexly written programming language. The only popular programming language that satisfies this condition is Python; it's the most recommended language for beginners due to its simple syntax, and it's also used in backend web development, AI, ethical hacking, and automation.

Picking Python as a starting point lets you enjoy your learning journey. The opposite is true; starting with a complex language like C++, despite its importance, it will make you struggle a lot, and maybe you will give up soon. However, some beginners prefer to start with C++ as a challenge to discover low-level concepts. Learning how to code is like a game. You enjoy the hard levels after you finish the easy ones, but some players start directly from the hard levels as a challenge. The recommended path for most people is to start with Python, then move to the next levels like HTML, CSS, JavaScript, SQL, PHP, C++, etc.

Top Recommended Free Sources to Learn Programming

Learning how to code doesn't require money from you; you can actually learn for free. The only requirements are internet, a computer, and a distraction-free environment. But the most important requirement is not physical; you need to be disciplined to achieve your goal, which is another topic.

To learn programming or programming languages for free, including Python and other languages, I recommend the freeCodeCamp website or its YouTube channel. It provides comprehensive programming tutorials for no cost. I won't limit you to one source. There are also CS50, MIT OpenCourseWare, and Khan Academy for solid tutorials, and the documentation websites for each programming language, like Python documentation and MDN Web Docs, as references.

Bottom Line

Start learning now; don't let AI evolution be your excuse to avoid learning. Start with your motivation: you want to make money, right? Pick a project that has a big potential to make money online, especially passive income, such as a SaaS website, a blog, a mobile application, a mobile game, etc., then search online to decide the best programming languages or frameworks for this project. Then start exploring tutorials and documentation related to those languages or frameworks, and finally start building your project. It's OK to stay blocked, maybe for an hour or even days, without any idea how to start or how to fix your current problem. That's totally fine; all programmers started like that. Don't be disappointed early. Fortunately, you are in the age of AI, where you can get an explanation for any problems instantly. Learning how to code is much easier now. Just start, and don't wait for the perfect moment; your perfect moment is when you actually start. The journey of thousands of steps starts with one step. Happy learning and thanks for reading.