Will Emotions Become the New Programming Language?
“Tools will come and go, but the programmer of feeling is the only one who remains.”
With this profound phrase, music producer Rick Rubin summarizes a truth that goes beyond music and extends into every creative field—including programming. Tools are merely shifting mediums, while the real value lies in the person who uses them to translate their feelings and ideas into something alive.
Since Alan Turing laid the foundations of what is now known as the theory of computation, programming began as a language for communicating with machines. Then came the first programming languages, like Fortran in the 1950s, which gave humans the ability to turn abstract ideas into executable instructions.
Since then, programming has evolved from manual, complex writing to intelligent systems, yet it remained mostly tied to those with advanced technical backgrounds.
However, this reality is now undergoing a radical transformation led by artificial intelligence tools, which are redefining programming as a creative act open to everyone—even those without technical expertise. Among these tools, a new approach is emerging known as “Vibe Coding” or “Programming by Feel”, a method that allows the creation of digital applications and services through intuitive interaction with AI, rather than rigid programming commands.
What is Vibe Coding, and Where Did It Come From?
The term “Vibe Coding” was coined by prominent engineer Andrej Karpathy, former head of AI at Tesla and one of the founding technical members at OpenAI. In February, Karpathy posted on X (formerly Twitter):
“This is a new kind of programming I call ‘Vibe Coding’—when you fully surrender to the vibe, embrace acceleration, and forget that code even exists.”
Through this lens, programming transforms from a strict technical activity into a process closer to artistic creation, where the user interacts with tools like ChatGPT, Copilot, or Claude using natural language descriptions. The AI then automatically generates the code.
This idea has resonated strongly, especially among non-specialist users who found themselves capable of building prototypes and real applications without needing to learn traditional programming languages. Developer Simon Willison noted:
“It captured a moment that resonated with a lot of people, because many developers were already using these tools and achieving amazing results.”
This new approach doesn’t require writing every line of code manually. It’s enough to describe a general idea or the “vibe” you want the application to have. From there, the AI transforms this vision into executable code.
Initially, the term “Vibe Coding” may have started as a Silicon Valley trend, but it quickly evolved into a genuine movement reshaping how software is developed. Instead of strict adherence to syntax and classical structure, a user can now simply describe a problem or concept in plain English, and the smart tool will translate it into functional code.
Karpathy joked:
“The latest programming language is English.”
This encapsulates the notion that successful programming now relies not just on coding knowledge, but on clarity of vision and the ability to express it.
He also shared his experience, saying:
“It’s not really programming—I just see things, say things, run things, copy-paste things, and mostly it works.”
This highlights the fundamental difference between the new intuitive approach and the traditional, tightly controlled method.
How Is AI Changing the Face of Programming?
This shift has opened the door for a much wider audience to enter the world of software development—even those who don’t know any programming language. Anyone with a clear idea or a vision for an experience can now build a prototype or even a complete product in collaboration with AI.
At the same time, this new style doesn’t exclude professional developers; rather, it provides them with tools to accelerate workflows and automate repetitive tasks, allowing them to focus more on design and creative direction. Routine tasks are now executed at incredible speeds, narrowing the gap between idea and implementation.
Thanks to these advancements, software engineering is expected to look radically different by the end of 2025, as AI becomes increasingly capable of understanding natural language commands and converting them directly into working applications.
How Does Vibe Coding Work?
“Vibe Coding” or “Programming by Feel” works through a close interaction between human guidance and AI-generated content. The user provides instructions, descriptions, or goals in everyday language, and AI tools translate this into code. The process is usually iterative and collaborative, following a pattern like this:
-
You describe what you want:
For example: “Create a simple web page that shows the current weather for a city entered by the user.”
This prompt is given to an AI coding assistant. -
AI generates the code:
Based on its training on large datasets of code, the AI tries to generate code that fulfills your request. It may produce HTML, CSS, and JavaScript automatically. Think of it as supercharged autocomplete, predicting the code based on your input. -
You review and guide:
After the AI provides the code, you test and evaluate it. Does the page behave as you imagined? If the first version has issues, you provide further instructions like: “Make the design more colorful and add error handling if the city is not found.” The AI then updates the code accordingly. -
Repeat as needed:
This loop (describe → generate → review → refine) continues until you’re satisfied. Thanks to the AI’s speed, you can go through multiple iterations quickly.
Modern AI tools show astonishing power in this context—not only writing code, but also helping to debug and enhance it. If a piece of code runs into an error, you can paste the error message into the AI and ask for help.
Karpathy noted that this method often solves problems instantly, with the AI reading the error, explaining it, and offering a fix.
On the other hand, with the “Vibe Coding” approach, there’s no need to memorize the exact syntax for loops or function details. You just express what you want in natural language, and the AI writes the code for you—as if you’re working with a highly skilled programming assistant.
Tools like GitHub Copilot have made this capability very clear. When given a programming problem described in plain English, Copilot can generate code to solve it. It can even turn English comments into executable code and auto-complete entire functions.
Similarly, “Vibe Coding” expands this concept to cover larger tasks, such as generating full software modules or basic applications from high-level descriptions