Vibe Coding for Teachers: Start Small, Build What You Need

When people hear “coding,” they often imagine something complicated and inaccessible—something for software developers or people with a computer science background.

AI is changing that. Vibe coding is a way of using ordinary language to describe what you want, then working with AI to create or change the code behind it. You don’t have to know how to write code to get started. You do need an idea, a willingness to experiment, and the patience to test what you make.

For teachers, that opens up an interesting possibility: we can try building digital tools that respond to the things we actually teach and the problems we actually face.

My first experiment

I first tried vibe coding about a year and a half ago. I made a simple choose-your-own-adventure story about the Stone Age. Students would read part of the story and make choices at certain points.

The result wasn’t great, and the process involved more friction than it does now. But it was exciting to see an idea become something people could click through. I left it there for a while, then returned to experimenting a few months ago. Since then, I’ve built several projects that I use in my work.

One wasn’t for students at all. When I took on a professional development coordinator role, I had a lot to keep track of: plans, goals, observations, meetings, activities, and hours. The information was spread across documents and different communication platforms. I wanted one place where I could see the overall picture, so I used AI to help me build a PD management tool.

It has made it easier for me to organize and track the work. The useful starting point wasn’t “I want to make an app.” It was “I’m finding it difficult to keep this information organized. Could a digital tool help?”

Building experiences for students

My other projects are closer to my work as a history teacher.

For a Grade 10 unit on prehistory, I built a digital Stone Age cave. Students enter their names and explore three sections: stone tools, fire, and cave art. They complete observations and interpretations, answer multiple-choice and open-ended questions, and finish with an archaeological investigation. I can then review their work in a teacher portal.

I also made an experimental archaeology simulation. Students enter a fictional institute and work through five cases, trying out the kinds of questions and tests an archaeologist might use. Students have used it and enjoyed it. Their feedback has also given me ideas for what to improve.

I’m now working on a larger project: a two-dimensional game set in the Indus Valley civilization. It’s my first game, and I’m building it in stages. It takes more time than a small activity, but the process feels much more approachable than I would have expected before trying it.

A digital activity needs a reason to be digital

Making something interactive doesn’t automatically make it a better learning experience.

One student told me that the experimental archaeology simulation involved too much reading and too many clicks. It didn’t feel as immersive as it could. That feedback made me think about what digital tools can offer beyond putting text on a screen. Could the activity use sound, movement, choices, images, or other interactions that would be difficult to create on paper? Would those features help students learn, or would they just be decoration?

Feedback also matters for the experience of using a tool. Students have pointed out things like text that is too small or an interface that feels awkward. Those details affect whether students can focus on the learning.

So the cycle isn’t just “ask AI to build it.” It’s build something, try it, gather feedback, and improve it. The educational purpose matters just as much as whether the tool works.

How teachers can get started

My main advice is to start small. If your first project is too ambitious, it’s easy to feel overwhelmed and give up. Pick something simple enough to finish and learn from.

You could start with:

• A short choose-your-own-adventure story for a topic you teach.

• A simple quiz or review game.

• A digital version of one activity you already use.

• A tool that organizes information you currently track across several places.

Before you start building, describe the problem. What is taking too long? What do students find confusing? What information is hard to organize? Then ask whether a digital tool would actually help. Some problems don’t need an app or website, and that’s useful to realize early.

Once you have an idea, ask the AI to help you plan the simplest version. Talk through what the user should see and do, one step at a time. Build the foundation first, then add features gradually. For a bigger project, work in phases and try each part before moving on.

You can use an AI tool that can create and edit code. I’ve used ChatGPT and Claude, but the important thing is to experiment with a tool that’s available to you and suited to the task. You can begin by describing what you want in everyday language. For example:

“I teach Grade 10 history. Help me plan a simple digital activity about Stone Age cave art. First, ask me questions about the learning goal and what students should do. Then suggest the smallest version we could build. Don’t write code yet.”

That last instruction can be helpful: get the idea and learning design clear before you ask for a working product.

Check the content and protect student information

AI can make mistakes, including confident mistakes. Check the content it generates, especially when students will use it to learn. For my Indus Valley game, I’m grounding the material in real academic sources and references rather than relying on AI to make things up.

Be thoughtful about student names, work, and other personal information, too. Follow your school’s policies and use approved tools and storage. Only collect the information you need. A login or teacher portal doesn’t automatically mean student data is secure, so check how access and storage work before using a tool with real student information. If you’re unsure, ask the appropriate person at your school.

Keep experimenting

Vibe coding has made it possible for me to try ideas I probably wouldn’t have attempted before. I’m not a developer, and I don’t have a computer science background. I’ve learned by making things, seeing what works, and changing what doesn’t.

You don’t need to start with a complicated app or a big game. Start with a small problem or a fun idea. Make the simplest version. Try it yourself, let someone else try it, and ask what would make it clearer or more useful.

The first version doesn’t have to be perfect. It just has to give you something to learn from.