What I Do
I build digital tools and web applications. Most of them start with a fairly simple question. Usually something along the lines of "Wouldn't it be useful if...?" or "I wonder if this could actually work?"
And that's often all it takes.
Sometimes the idea comes from a real problem. At MIKS Ressurssenter, where I work with entrepreneurship and training, I often come across things that could be done a little differently. Other times it's just something that pops into my head. Like an app that helps a group of friends agree on what movie to watch. Or a website where rain falls on a pond and three koi occasionally swim past.
Now, here's the thing about these ideas. They tend to grow.
I'll start with something relatively straightforward, and before long I'm thinking about how people will actually use it, what might confuse them, what happens when something goes wrong, and whether there's a better way to approach the whole thing.
I enjoy that part enormously. Taking something that's still a little fuzzy in my head and gradually figuring out what it could become.
And this is where Codex comes in.
I started using ChatGPT back in 2022, with practically no programming experience. These days, Codex is part of my everyday development process. And the way I work with it has changed quite a bit over time.
In the beginning, I was mostly interested in getting things to work. Now I spend much more time figuring out what we should build in the first place.
Before starting a larger project, I like to challenge the idea. Sometimes I'll have Codex ask me a whole series of questions, using a method called grill-me, until we've uncovered the assumptions, decisions and little complications hiding beneath the surface. For more complex projects, I use Wayfinder to map out the work before we begin.
Then we start building.
Usually in small steps. Codex helps with the implementation, tests and technical decisions. I focus on the product, the experience and whether we're actually solving the problem we set out to solve.
There's a lot of back and forth. I'll test something, notice a detail, change my mind, ask another question, or decide that an entire section needs a different approach. Sometimes we end up revisiting something we thought was finished.
And every now and then, a seemingly tiny detail turns into a rather large project of its own.
(If you've spent any time watching the fish on the Raindance homepage, you've seen the result of a few of those decisions.)
One thing this whole experience has taught me is that being able to build something changes the way you think about ideas. Things that once seemed far beyond my technical abilities are now things I can actually explore, test and sometimes turn into working products.
That still fascinates me.
I also enjoy seeing how other people approach similar problems. I've learned an incredible amount from developers who share their workflows, explain their decisions and talk openly about the things that went wrong. Whenever I discover something useful myself, I like to pass it along.
Some of my projects are developed for organisations, others are independent experiments. And I'm always interested in meeting people who are working on something interesting, especially if there's an opportunity to learn something new or build something together.
That's really what Raindance is about. A place for the things I build, the ideas I explore, and whatever comes next.