What I Learned Teaching 100 People To Code with AI

Last month I launched a cohort-based course through Every: How to Build an AI Chatbot. The idea was to teach people how to make a GPT-4 chatbot by programming with ChatGPT over the course of a month. 

AI is a hot topic, and my articles on it have done well. I felt like I had something genuinely important to teach. So, I had high expectations for the course. But I was completely unprepared for the level of response we got.

We got a trickle of signups in the first hour. But over the course of the next few days—the trickle kept coming, and coming, and coming. I watched the number of registered students go from 5 to 10 to 25.

By the 80th student, we shut off signups. We gave out 20 scholarships to students who couldn’t otherwise afford a seat, and I started to prepare to teach. My terror that no one would show up was replaced by something else: Will anyone like it? More importantly, will they truly learn something? Will it be valuable?

Over the last month, I’ve had the absolute honor of watching almost 100 people get their feet wet building with AI. There were big WOW moments—like when students got their first chatbot running in the initial lecture, or when they suddenly understood an AI term they’d been hearing about forever like “embeddings”. There were also some bumps and bruises along the way—it turns out, AI is still unreliable at programming and that I personally have some ways I can improve as an instructor.

But the incredible thing about courses like this is that if you’re doing it right, you learn just as much as your students do. That was certainly the case for me this month. It was a huge pleasure to help so many people get their feet wet building with AI—and it was a crash course on being a programming teacher in this new world.

Here are a few things I learned.

AI is a cheat-code level powerup for creativity

I went into this course believing that AI is a video game cheat code powerup for creativity that can take people who are only marginally technical and turn them into builders who can suddenly bring their ideas to life. All they need is a little push.

I believe that even more fervently now. I watched people in this course go in with only basic technical skills and have immediate WOW moments that stoked their creativity and sense of agency. In the first class everyone made a basic conversational chatbot that looked like this:

At the beginning of class, I demoed the bot and stepped through sample code to explain how it worked. Then I live-coded the bot from scratch using ChatGPT so students could see how to properly prompt it to get the code they needed. Then I set them loose to use ChatGPT themselves to code their own bot. It was risky—the class included people from all different programming skill levels. But it turned into a huge moment for a fair number of them. Here are a few things people said in the chat as they were working through the exercise:

“It worked!!! I'm in shock right now ;)”

“It's working! It feels like cheating—I barely did anything.”

Programming with AI for the first time is like that. It sort of feels like that moment in The Matrix where Morpheus downloads martial arts skills into Neo’s brain. Neo wakes up and suddenly knows kung fu. 

Getting people to that initial feeling of creative agency with AI is great because it is real. They suddenly realize that they can do much more than they thought they could if they use AI to power their workflow. 

But it’s also good for a deeper reason: it helps motivate understanding.

AI makes learning to code way less intimidating and abstract