Vibe Coding: The Era Where Everyone Is a Programmer Has Really Arrived
One afternoon, I said something to my terminal window: "Help me build a bilingual Chinese-English personal website with physics engine animations."
Then I hit Enter.
I don't know frontend development at all. I'd never written React, and Next.js routing was completely foreign to me. But that same afternoon, the website you're looking at right now—75 TypeScript files, 29 React components, a draggable tag wall built with the Matter.js physics engine, and an AI chat assistant connected to a large language model—was literally "talked" into existence.
All the code was written by Claude Code. I didn't write a single word of it.
This is called vibe coding. A year ago, most people hadn't even heard the term.
What Exactly Is Vibe Coding?
On February 2, 2025, Andrej Karpathy posted a casual tweet on X:
"I just see stuff, say stuff, run stuff, and copy paste stuff, and it mostly works."
He called this new way of programming vibe coding. You no longer write code line by line; instead, you tell the AI what you want in everyday language, then watch it build things. When you hit an error, you toss the error message at it, and it usually fixes it. The codebase grows to a size you can't even read, and you simply don't look at it anymore.
Karpathy himself said it was just a spontaneous shower thought. But it seemed to name something many people were already doing without knowing what to call it.
You probably noticed what happened next—Collins Dictionary named vibe coding its 2025 Word of the Year, and Google searches for the term skyrocketed by 6,700%. A casual tweet became the name of a movement.
To put it simply, vibe coding boils down to one sentence: you don't need to know code; you just need to be able to clearly say what you want.
From Completion to Conversation: Three Leaps in AI Programming
AI helping people write code isn't new. But the past few years have seen three completely distinct leaps. Understanding the differences between them is key to grasping what's actually new about vibe coding.
The first step was code completion. When GitHub launched Copilot in 2021, programmers got excited—it could guess what you were going to write next based on half-finished code. Much like predictive text on your phone, it made typing faster, but you still had to know what to type. Non-programmers staring at Copilot were just as lost as before.
The second step was code generation. After ChatGPT took off in 2023, you could describe a requirement in Chinese and it would spit out a code snippet. This was a big leap from Copilot, but you still had to understand the code, know where to put it, and fix bugs yourself. It was like having a fast but somewhat unreliable intern—helpful, but you had to watch them closely.
The third step is vibe coding. Starting in early 2025, tools like Claude Code and Cursor could read your entire project codebase, create files, modify files, run tests, fix bugs, and even find workarounds when stuck. You went from being the person writing code to the person making requests. In essence, you became the product manager, and the AI became the engineering team.
The difference here isn't one of degree; it's one of kind. From "AI helps you write code" to "you tell the AI what you want," there's a threshold crossed. As I discussed in my previous article, AI isn't the next Copilot—it's a category jump.
How High Is the Adoption Rate of AI Programming?
Data from last year's Y Combinator winter batch surprised a lot of people. YC CEO Garry Tan said that 25% of startups in that cohort had codebases that were 95% AI-generated.
YC is a top-tier global startup incubator; these are people seriously raising funding and shipping products.
During the same period, Stack Overflow's developer survey showed that 84% of programmers use AI tools in their daily work. At LlamaCon, Nadella said that 20% to 30% of Microsoft's own codebase was written by AI.
Jensen Huang put it in a way that I think captures this most accurately:
"Everyone is a programmer. The new programming language is human language."
Two years ago, this sounded like a vision. In hindsight, it reads more like a plain description of what was already happening.
My Personal Experiment
Back to the website I mentioned at the beginning. To be honest, I wasn't very confident when I decided to build it.
The tech stack sounds intimidating on paper: Next.js 16, React 19, TypeScript strict mode, Tailwind CSS v4, Velite + MDX content system, Matter.js physics engine. But I knew nothing about any of this—not being modest, I literally understood none of it.
All I did was talk to Claude Code. All the code was written by it; I didn't touch a single word.
The particle animation on the homepage—floating light dots that follow your mouse when you refresh—is backed by 187 lines of Canvas rendering code that handles device pixel ratio adaptation, dark mode switching, and even detects the user's "reduce motion" preference. I know nothing about Canvas programming, but I could describe the visual effect I wanted to the AI and iterate round by round.
The draggable physics tag wall is even more interesting. Under the hood it runs on the Matter.js physics engine, where each tag has gravity, friction, and restitution coefficients, and bounces back automatically if it falls off the screen. All 262 lines were written by Claude Code; I didn't type a single one. But every line exists because I said something like, "If a tag falls down, it should bounce back."
Then there's the AI chat assistant—a complete 299-line chat interface connected to a large language model API, supporting streaming output, with a WeChat-style UI in Chinese environments and a WhatsApp-style UI in English environments.
Domain configuration, DNS setup, Vercel deployment—all things I completely didn't understand—were also handled by Claude. I didn't even know what DNS was; I just watched Claude clicking around in the browser and then telling me, "Done."
The whole project took one afternoon. If I had to learn these tech stacks from scratch and write it myself, I wouldn't have even started, because I had no idea where to begin.
I'm not watching this transformation happen from the sidelines. I'm using it to build houses—without moving a single brick myself.
Is AI-Generated Code Actually Reliable?
If I only wrote about the good stuff up to this point, this article would turn into a paid promotion. It's not that simple.
In July 2025, METR published a rigorous controlled experiment: they had experienced open-source developers (averaging 5 years of experience and 1,500 commits) use AI tools for development tasks on their own projects. The result: with AI, these people actually became 19% slower. Even more intriguing, the developers themselves felt that AI made them 20% faster—perception and reality were completely inverted.
Another data point comes from CodeRabbit's analysis at the end of 2025: code with AI involvement had 1.7x the rate of serious issues compared to purely human-written code, and security vulnerabilities were 2.74x higher.
These numbers are real, and there's no point in avoiding them.
But I think they're actually describing two different things. The METR experiment measured "experts doing work they already know well"—it's like asking a race car driver to drive while explaining every turn to the passenger; of course he'll slow down. But the point of vibe coding was never to make the race car driver faster. It's to let people who could never get on the road start driving.
As for security vulnerabilities, this shows that AI-written code does need human review before hitting production. But think about it: without vibe coding, many of these projects wouldn't exist at all. The right question isn't "Is AI-written code perfect?" but "Is it good enough to build things that were previously impossible?" For personal projects, prototype validation, and internal tools, the answer is clearly yes. Later, I used the same approach to write 300,000 lines of code in 10 days, then deleted all of it—that experience taught me that code is liability, not asset.
These problems are real. But tools are iterating rapidly; weaknesses from six months ago may already be fixed today. The direction is right; the road is still under construction.
Code Is Just the Beginning
If you're still thinking vibe coding is just "making websites without learning programming," you're underestimating it.
Following the thread from my previous article: when AI can directly convert capital into productivity, what exactly is the conversion mechanism in the middle? I believe the answer is code.
The digital world runs on code. Every app on your phone, every website you use, every online transaction—all of it is powered by running code. Code is the universal interface through which humans control the digital world.
And vibe coding means AI has mastered this interface.
When AI can write code reliably, it can build software. When it can build software, it can automate almost any digital task—booking flights, managing schedules, analyzing reports, building websites, calling APIs. These tasks are all essentially variations of "write a program and run it."
This is why the AI Agents I discussed in the previous article are so worth paying attention to. For an agent to do things for you in the digital world, it needs to be able to operate that world. How? By writing code, calling APIs, reading and writing files. Vibe coding gives the agent this capability. Or put another way: vibe coding is the interface between you and the AI agent—you express intent in natural language, and the AI turns it into reality with code.
The previous article said that "capital can bypass labor and convert directly into productivity." Vibe coding is the bypass.
Your First Step
If you've read this far, you might have two voices in your head. One says, "This is so cool," and the other says, "But I don't know how to program."
The good news is that the second voice is describing exactly the problem vibe coding solves. You don't need to know programming. You just need to be able to talk and type.
I use Claude Code, a terminal-based AI coding tool from Anthropic. Installation takes just one command.
For macOS or Linux users, open your terminal:
curl -fsSL https://claude.ai/install.sh | bash
For Windows users, open PowerShell:
irm https://claude.ai/install.ps1 | iex
Once installed, open a terminal in any folder, type claude, and try saying something like: "Help me write a small BMI calculator."
See what happens.
What you just did would have required a computer science degree five years ago. Three years ago, you would
Originally published at https://guanjiawei.ai/en/blog/vibe-coding