All posts

Start here · Curious about tech · Explainers

What is coding, actually? Simpler than anyone told you.

You have heard for years that coding matters, that people should learn it, that it pays well - and probably nobody ever explained what it actually is without burying you in jargon. This page is that explanation: what code really is, what it looks like (you can already read some), what coders do all day, and how AI just changed the answer. No prior knowledge assumed, ever.

Deric YeeDeric Yee Updated 25 August 2026 9 min read
A laptop showing code - less mysterious than it looks

The whole idea in one paragraph

A computer is a machine that follows instructions - insanely fast, and insanely literally. It does not guess, infer, or fill in blanks the way a human colleague would. Coding is writing instructions precise enough for that literal-minded machine to follow- that is the entire concept. When you order on Grab, somewhere a list of instructions someone wrote is running: check the address, find nearby drivers, calculate the fare, send the notification. The instructions are written in special languages (Python, JavaScript - human-invented, far smaller than English, designed to be unambiguous), and a “program” or “app” is just a big, organised pile of them. Everything else you have heard - algorithms, software, developers - is vocabulary sitting on top of this one idea: people writing very precise instructions for very fast, very literal machines.

And here is the part that surprises people: you have probably already done a version of it. If you have ever built an Excel formula- =IF(B2>150, 0, 8) to set a delivery charge - you have written a computer instruction with conditions and logic. That is programming, in a spreadsheet costume. Real code is the same kind of thinking with more room to build. See for yourself - this is genuine code, and you can already read it:

your-first-look.js - real code, readable on sight
// This is real code - and you can probably already read it.
// It checks a shopping cart and decides the delivery fee.

if (cartTotal > 150) {
  deliveryFee = 0            // free delivery over RM 150
} else {
  deliveryFee = 8
}

message = 'Your total is RM ' + (cartTotal + deliveryFee)

That’s it. Not hieroglyphics - English-ish words, some symbols, and a rule: if the cart is above RM 150, delivery is free. Real professional code has more of it and it is organised more carefully, but the texture is exactly this. The mystique around coding survives mostly because people never get shown this page-one view.

So what do coders actually do all day?

Less typing than you imagine, and more thinking. A working developer’s day is mostly: translating fuzzy human wants into precise steps(“customers should get a reminder” becomes: which customers, when exactly, what if they already paid, what if their number changed?); detective work(something worked yesterday and doesn’t today - finding out why is the single most time-consuming activity in the profession, and the people who enjoy it as puzzle-solving thrive); reading and improving existing instructions (most code work happens inside things already built, like renovating rather than building from empty land); and, in 2026, directing and checking AI- modern developers have AI assistants write a lot of the first-draft code, and spend their skill on specifying what to build and verifying it is right. If that sounds less like “maths genius” and more like “precise, patient problem-solving” - that is accurate, and it is why such a wide range of people turn out to be good at it.

Why everyone suddenly cares that you learn this

The honest answer has two halves. The old half: software runs everything, so people who can build and fix it have been paid well for decades - in Malaysia, junior developers who can work the modern way start at RM 6,000–9,000 a month, roughly double many fresh-grad office salaries (the full data is in our hiring report). The new half: AI made basic instruction-giving a skill for everyone, the way Excel became a skill for everyone in the 1990s - our essay AI is the new Excel tells that story with the numbers. You do not have to become a professional developer for coding literacy to pay off; understanding how software thinks is becoming the general workplace advantage that spreadsheet literacy was for your parents’ generation. And if the curiosity runs deeper than literacy, the door happens to be wide open right now.

Try the real thing tonight - free, both ways

Reading about coding is like reading about swimming; the water is where you find out. The gentlest version: our free tutorials - the entire beginner course published openly, over a hundred bite-size lessons, readable with no account and no email, starting with coding in your browser with nothing installed. The fifteen-minute version:open ChatGPT or Claude and type “build me a simple webpage that splits a dinner bill between friends - one file I can open in my browser”, save the result as a .html file, and double-click it. You just made a computer do your bidding; notice how that feels. The proper taste: the free trial - one signup, no card - gives you real guided projects on Sigmo with a patient AI tutor, plus a live session with a human instructor who answers the questions this article raised. If you want to know what building leads to, what can I build with AI shows the ladder - and if you’re wondering which corner of tech fits your personality, which tech career fits me maps the options in the same plain language as this page.

FAQ

  • What is coding, in the simplest terms?

    Coding is writing instructions for a computer - that is the entire idea. Computers are extremely fast and extremely literal: they will do exactly what you say, millions of times a second, but only if you say it precisely. Code is just the written form of those precise instructions. Everything digital you touched today - WhatsApp, your banking app, Grab, the traffic lights’ timing system - is a pile of such instructions that someone wrote. The closest everyday experiences: writing a very exact recipe for someone who cannot improvise, or building a formula in Excel - which is genuinely a form of programming most office workers already do without calling it that.

  • Is coding like maths?

    Far less than school led you to believe. Day-to-day coding is closer to writing and logical planning than calculation: you spend your time naming things clearly, breaking a problem into steps ("first check if the user is logged in, then show their orders"), and figuring out why step three isn’t doing what you expected. The maths that appears is mostly arithmetic - totals, percentages, dates. Heavy mathematics only appears in specialised corners (game physics, machine-learning research) that most working developers never touch. Plenty of professional developers, including graduates of ours, were "bad at maths" in school; what they were actually good at was patient, structured thinking - which is the real skill.

  • How is coding different now that AI exists?

    AI changed the typing more than the thinking. Modern AI tools can write code from a plain-English description - so the mechanical part of coding (remembering syntax, typing it out) has become dramatically easier. What AI did not change: someone still has to decide precisely what to build, check that the AI’s code actually does the right thing, and fix it when it doesn’t - and that judgement still requires understanding how code works. Think of it like calculators and accounting: calculators didn’t end accountants, they ended manual arithmetic and made the judgement part of the job more valuable. Learning to code in 2026 means learning that judgement, with AI handling more of the typing.

  • Can a normal person actually learn coding?

    Yes - and the evidence is not motivational-poster talk. The people who become developers through modern programmes are overwhelmingly former teachers, accountants, retail managers, fresh grads from unrelated degrees - normal people who put in structured hours. The honest requirements are patience with being confused (everyone is, at the start), consistency (steady hours beat bursts), and tolerance for fixing your own mistakes - the actual daily texture of the work. What is NOT required: youth, a science background, or genius. The realistic timeline from zero to employable is measured in months of focused effort (400-600 hours), not years - and you can test whether you enjoy it in a single free evening.

  • How can I try coding tonight, for free, without installing anything?

    Two genuinely free ways. Fastest: open an AI assistant (ChatGPT or Claude) and type: "Build me a simple webpage that [something personal - splits a dinner bill, counts down to my holiday]. Give me one file I can open in my browser." Save what it gives you as a .html file, double-click it, and you have made the computer do your bidding - the core feeling of coding, in fifteen minutes. More structured: the Sigmaschool free trial (one signup, no credit card) gives you real guided mini-projects on Sigmo with an AI tutor that explains everything in plain language, plus a live session with a human instructor. The first is a spark; the second tells you whether the spark is a career signal.

Back to all postsNext: how apps are actually made →

The water is where you find out.
Fifteen curious minutes, zero ringgit.

Real guided projects, a patient AI tutor, and a live human instructor - the free trial is built for exactly the curiosity that got you to the end of this page. One signup, no card.

Try coding free tonightWhich tech path fits you?