Basics
What vibe coding is, how AI models work and where their mistakes come from.
Start here
What vibe coding really is and when to use it
The term sounds like a joke, yet it describes a real shift in how people work with code. We explain where it came from, what it means in practice and where its usefulness ends.
Read the article ›Path for beginners
Five texts in a sensible order: from the idea to safe publishing.
- 1What vibe coding really is and when to use itBasics · 7 min read
- 2What artificial intelligence and large language models are, in plain wordsBasics · 6 min read
- 3Prompt engineering for everyone: role, context, format and examplesPractice · 7 min read
- 4Your first project step by step: from an idea to a working pagePractice · 7 min read
- 5Risks of AI-generated code: a checklist before deploymentSecurity · 8 min read
Sections
All articles
- What vibe coding really is and when to use itBasicsThe term sounds like a joke, yet it describes a real shift in how people work with code. We explain where it came from, what it means in practice and where its usefulness ends.
- What artificial intelligence and large language models are, in plain wordsBasicsAI, machine learning, LLM: these terms often get mixed up. We sort out the concepts and explain why a language model can write code as well.
- Tokens, context and hallucinations: why the model gets it wrongBasicsWhy does a model "forget" the start of a conversation, confidently say untrue things and have limits? Three concepts explain most of these puzzles.
- Vibe coding versus "real" programming: do you need to know how to code?BasicsCan anyone build an app thanks to AI? Up to a point, yes. We look at where the line lies and how to move it.
- How to write prompts for code that get you useful resultsPracticeThe difference between "write me an app" and a well-described task is often the difference between chaos and working code. Five habits that change it.
- Prompt engineering for everyone: role, context, format and examplesPracticeA good instruction is not a magic spell but a handful of repeatable elements. We show a template that works for most tasks.
- Your first project step by step: from an idea to a working pagePracticeInstead of building a huge app, start with a small, working version. Here is a plan from an idea to your first published page.
- Debugging with AI: how to fix errors without panicPracticeAn error is a normal part of the work. We show a routine that gets you out of the "the model keeps fixing it, yet it still does not work" loop.
- Git for beginners: your safety netPracticeA model can "fix" working code so that it stops working. Git lets you go back to the last good version in a few seconds.
- AI coding tools: how to compare them instead of chasing rankingsToolsThe offering keeps changing, so instead of yet another list of names we propose an evaluation framework that is useful whatever happens to be fashionable.
- AI agents: when the model carries out the steps itselfToolsA chat answers a question. An agent is given a goal and plans the steps, runs tools and checks the results on its own. Convenient, but it needs supervision.
- Risks of AI-generated code: a checklist before deploymentSecurityCode that works is not necessarily secure or maintainable. We review the most common traps and propose a list of questions to ask before every deployment.
- Privacy and data: what not to paste into AISecurityPasting a code snippet takes half a second. The consequences of pasting a key or customer data can last much longer. Rules worth keeping in mind.
Nothing found. Try a different word.
Glossary
- LLM (large language model)
- An AI model trained on huge collections of text and code that generates responses to instructions. It is the engine behind most AI assistants.
- Prompt
- An instruction or message given to an AI model. The clearer and fuller it is, the better the result.
- Token
- A piece of text in which the model "reads" and counts. Limits and often costs depend on the number of tokens.
- Hallucination
- An answer that sounds convincing but is made up or wrong, for example a library function that does not exist.
- AI agent
- A model with access to tools that plans steps on its own, carries them out and checks the results.
- MVP
- Minimum viable product: the smallest version of a product that already works and lets you test an idea.