Vibe coding in education: your own digital learning tools without programming knowledge

Every educator eventually faces the same situation: there is no digital tool that does exactly what they need for a particular topic or group of learners. Existing tools are too complex, too simple, or simply not designed for that purpose. Most teachers then either settle for a compromise or drop the idea altogether. In a post on the Hungarian Újpedagógia blog, István Tóth suggests a third option: vibe coding.

What is vibe coding?

Vibe coding refers to describing in natural language what you want, while an AI-powered development tool turns that description into a working application. It requires neither a programming background nor years spent learning to code. What it does require is a well-defined pedagogical problem and a little patience.

The post draws an important distinction between simply asking AI to do something on your behalf and using it as a creative partner. In the first case, technology turns the educator into a passive consumer. In the second, the educator becomes a designer and decision-maker. Using a digital tool does not automatically make a process digitally pedagogical. With vibe coding, however, the teacher does not just receive a finished product: they actively shape, test and improve it, and keep ownership of the process throughout.

Why it matters for teaching

Ready-made digital platforms are convenient, but they have one basic limitation: someone else has decided what they are for. When educators build their own tools, they can tailor them to a specific class, subject and learning goal, rather than adapting their teaching to an existing template.

The process itself is also a learning experience. Educators learn how to break a pedagogical idea down into concrete, achievable steps and how to phrase a request so that it produces a usable result. This kind of structured thinking, based on design, testing and iteration, is useful well beyond technology and applies to any area of pedagogical development.

Getting started

According to the author, the biggest obstacle is not technical but a matter of mindset: educators need to believe that a clearly formulated idea can really become something tangible. The post recommends a step-by-step approach:

• Start simple. Build a basic version with a single function, test it, then expand it gradually.
• Use a familiar tool. Starting with an AI tool you have already practised with reduces misunderstandings and unnecessary debugging.
• Share it with students. The final step is to make the app available beyond your own computer. An idea only becomes a pedagogical tool once it reaches learners.

Lessons from practice

Újpedagógia applies this approach in its in-person training course for teachers, run over four sessions. Participants start by writing their first development prompt and leave the first session with a working app of their own. They then make it more stable and expandable, add motivating features such as saving points and progress, and finally publish it as a live web application that can be shared with anyone.

The key lesson was not technical. The participants who got furthest were those who knew exactly which learning problem they wanted to solve and then looked for the right tool, not the other way around.

AI as a creative partner

Discussions about the future of digital pedagogy often ask how educators can remain active, critical participants when using technology, rather than passive adopters. Vibe coding is one possible answer: not because every teacher needs to become a developer, but because it shows that the best pedagogical use of AI is as a partner that thinks with educators, not one that works in their place.

Source: When the teacher idea becomes an app: thoughts about vibe coding; Post of István Tóth (8 September) 2026.)