{"id":7084,"date":"2026-08-19T15:30:45","date_gmt":"2026-08-19T13:30:45","guid":{"rendered":"https:\/\/oneline.ch\/en\/?p=7084"},"modified":"2026-08-19T15:30:46","modified_gmt":"2026-08-19T13:30:46","slug":"how-does-vibe-coding-work","status":"publish","type":"post","link":"https:\/\/oneline.ch\/en\/how-does-vibe-coding-work\/","title":{"rendered":"How Does Vibe Coding Work? From Prompt to Code"},"content":{"rendered":"\n
\n

In Brief<\/strong><\/p>\n\n\n\n

Vibe coding works through a cycle of prompt, code, and correction: a task is described in plain language, an AI generates the code, the result is run, and adjusted again through a new prompt if needed. This process is often enough for quick prototypes; for growing or multi-person projects, a more controlled form of AI-assisted development comes into play.<\/p>\n<\/div><\/div>\n\n\n\n

To answer the question “How does vibe coding work?”, we first need to place the principle of this strongly AI-assisted approach to software development in context. Here, system requirements aren’t formulated directly in code, but in natural language, which an artificial intelligence then translates into program code. The typical process is simple: a task is described, the AI generates code, the result is run, and it’s then improved through new prompts. This cycle of prompt, code, and correction<\/strong> repeats until the desired function is achieved.<\/p>\n\n\n\n

In the narrow sense, the focus isn’t on reading the code in detail. What matters first is whether the application does what’s expected. For simple experiments and quick prototypes, this approach can bring a lot of speed. In larger projects, however, a more controlled form of AI-assisted development<\/strong> often comes into play.<\/p>\n\n\n\n

How Does Vibe Coding Work in Practice?<\/h2>\n\n\n\n

A First Prompt as an Example<\/h3>\n\n\n\n

At the start, there’s no programming language, only the desired result<\/strong>. Instead of creating functions, variables, or data structures yourself, you describe the task in a prompt.<\/p>\n\n\n\n

A simple example would be:<\/p>\n\n\n\n

\n

\u00abWrite a Python function that reads a CSV file and prints all the records it contains.\u00bb<\/p>\n<\/blockquote>\n\n\n\n

Anyone who lets AI generate code this way gets a first draft directly from the language model (large language model<\/strong>, LLM). The LLM generates code based on the described task and the patterns it knows from its training. The resulting program code is then run.<\/p>\n\n\n\n

If everything works as expected, the next step follows. If an error occurs or a function is missing, the result is described again<\/strong>.<\/p>\n\n\n\n

The Vibe Coding Loop in 6 Steps<\/h3>\n\n\n\n

The typical vibe coding loop<\/strong> consists of the following steps:<\/p>\n\n\n\n

    \n
  1. Describe the goal or desired function<\/li>\n\n\n\n
  2. Have the AI generate code<\/li>\n\n\n\n
  3. Run the generated code<\/li>\n\n\n\n
  4. Observe the result and error messages<\/li>\n\n\n\n
  5. Formulate corrections or new requirements<\/li>\n\n\n\n
  6. Test the revised code again<\/li>\n<\/ol>\n\n\n\n

    So what does vibe coding really come down to? The path to the result runs through an ongoing dialogue with the AI<\/strong>. Writing individual lines of code by hand fades into the background.<\/p>\n\n\n\n

    The name for this approach was coined in February 2025<\/strong> by AI researcher Andrej Karpathy. We’ve summarized the origin, definition, and limits of the term in a separate article<\/a>.<\/p>\n\n\n\n

    How Does a Prompt Become a Working Application?<\/h2>\n\n\n\n

    How does AI actually write code in practice? As an example, let’s look at building a small task planner. The first prompt might simply ask for a web application in which tasks can be entered and displayed.<\/p>\n\n\n\n

    The AI generates the necessary LLM code. Once it’s running, it may turn out that tasks can be added but not deleted<\/strong>. Instead of searching through the program code, a new instruction follows:<\/p>\n\n\n\n

    \n

    \u00abAdd a button to each task so it can be deleted.\u00bb<\/p>\n<\/blockquote>\n\n\n\n

    The next adjustment works. Now it becomes clear that all entries disappear once the application is closed<\/strong>. The next prompt could be:<\/p>\n\n\n\n

    \n

    \u00abSave the tasks so they’re still there after a restart.\u00bb<\/p>\n<\/blockquote>\n\n\n\n

    This is how the application grows step by step. New requirements emerge directly from the visible result.<\/strong> If an error message appears, it can simply be handed back to the AI. The model then tries to identify the cause and generate improved LLM code.<\/p>\n\n\n\n

    How Do You Write Good Prompts for AI-Generated Code?<\/h2>\n\n\n\n

    Anyone who wants to generate code with AI should describe the desired result as clearly as possible<\/strong>. A vague prompt leaves the model a lot of room for interpretation. For early experiments, that may be enough. Once specific functions are required, an efficient development process calls for more precise input.<\/p>\n\n\n\n

    \u00abCreate a to-do list\u00bb describes only the basic idea<\/strong>.<\/p>\n\n\n\n

    Much clearer would be:<\/p>\n\n\n\n

    \n

    \u00abCreate a simple to-do list as a web application. New tasks should be addable, markable as done, and deletable. Entries should still be there after a restart.\u00bb<\/p>\n<\/blockquote>\n\n\n\n

    How does AI actually write code based on input like that? The model tries to translate the described functions into a technically feasible structure<\/strong>. The clearer the goal and behavior are formulated, the smaller the room for wrong assumptions. Anyone who builds prompts systematically<\/a> also reduces the number of correction rounds.<\/p>\n\n\n\n

    Corrections also benefit from precise input. Instead of \u00abthat doesn’t work\u00bb, a precise description of the problem<\/strong> gets you to the goal far more reliably.<\/p>\n\n\n\n

    What’s the Difference Between Pure Vibe Coding and AI-Assisted Development?<\/h2>\n\n\n\n

    In the original, narrow sense of vibe coding, the generated program code stays largely in the background<\/strong>. Control focuses on the visible result. If something doesn’t work, the feedback goes back to the AI, which creates the next version. The approach delivers usable results quickly, but it isn’t always the right choice.<\/p>\n\n\n\n

    AI-assisted development goes a step further. Here, too, artificial intelligence is used to generate AI code and quickly produce a working application. Specialists, however, additionally check how the solution is built technically.<\/strong><\/p>\n\n\n\n

    The difference can be summarized like this:<\/p>\n\n\n\n