The Backstory of JikiHi everyone!

I've spent the last 2 years building out Jiki , starting with an idea, running a bootcamp to test my thinking, and then building out a (what I hope is kickass) product with Aron and Nicole.

I wanted to tell you about the journey. Why Jiki exists, why we invested our time and energy into it, and where I hope it will lead.

Before Jiki, there was Exercism

For those who don't know, I've spent the last decade building Exercism . Exercism is a platform that helps developers deepen their skills and learn new programming languages. It's been a huge collaborative effort with thousands of people committing code, hundreds of maintainers building out language tracks, and thousands of mentors supporting other developers. Millions of people have used it to level up their coding, and it's something I'm really proud of.

But while Exercism was designed for developers who already know how to code, we kept attracting total beginners . In fact, at last count, over 500 people a day are signing up to Exercism who've never written a line of code before. And Exercism really doesn't work for them as it wasn't built for them, so we send them away.

And that sucks.

My motivation with Exercism has always been about social mobility - helping people who need that help the most. I believe that programming is a core skill that almost everyone should learn. For some it can be a career, but for everyone it teaches critical thinking, problem solving, and gives you a skillset that helps in almost any digital work. In short, I believe programming genuinely changes lives and opens doors that might otherwise stay closed. So watching these beginners arrive, full of enthusiasm, and then having to tell them "sorry, this isn't for you yet" felt like a failure on my part.

I started exploring the "learn to code" landscape to understand what was out there for them. I talked to lots of the beginners signing up and started to explore the resources out there. And as I tried those resources, I realised something was missing. Most of them don't get to the heart of what I believe programming actually is: fun problem solving .

Instead, they give you videos to watch, quizzes to take, maybe a little coding exercise, and a certificate at the end. They teach you syntax and theory, but they don't really get you making things. And I think that's why so many people give up. It's not that coding is too hard - it's that the way they're being taught doesn't feel like coding at all.

So I decided to do something about it.

Jiki's Journey to fruition

Basing Jiki on my experience

When I think about how I learned to code, it wasn't through courses or tutorials. I started at eight years old, long before I had the internet. I learned by building games . I'd turn whatever my childish mind was obsessed with at the time (there was lots of Star Trek and wizards) and turn it into some random game. Then as I got older I made websites and little tools for myself - whatever I found interesting. I didn't have a curriculum. I just made stuff, got stuck, figured it out, and made more stuff.

That experience has shaped everything about how I think about learning. I believe people learn to code by coding - not by watching someone else do it, not by answering multiple choice questions, not by writing a line of code to "finish" an exercise, but by actually writing real programs and solving problems.

That's why Jiki is built around projects from day one . You're not doing tiny five-line exercises for months on end. You're building things - games, animations, tools - and writing dozens, then hundreds of lines of code in your first couple of months. It's challenging, but it's the kind of challenge that makes you feel like you're actually becoming a programmer.

The Bootcamp

In January 2025, I ran a Bootcamp for 1,000 students. They learned a programming language that I wrote specifically for the bootcamp (called JikiScript). Each week they had about 3 hours of live teaching, then had some exercises to solve, then had a 3 hour "Labs" session, where we looked through the exercises together, they asked questions etc.

It was a really fun experience and I learned tons from it. It was also wildly intense as I was writing a programming language, exercises and lessons plus doing 6 hours of live streaming and dozens of hours of support each week - a huge thank you to the volunteers who helped me out with it!!

Crucially what I learned is that the pace I created was way too fast. I made something I thought would be gentle and easy, and it was still waaaaaaay too fast and hard for people. So I've taken that into Jiki, slowing things down even more - adding more exercises that go at a slower pace. If you're finding it easy, you'll shoot through those extra exercises, but if you're struggling, they'll (hopefully) be life-savers!

Removing the foot-guns

I strongly believe that learning to code is made a lot harder as you also have to learn a programming...