1995: The Beginning
I founded WebHouse ApS in 1995. Denmark's first web bureau, though we didn't call them that yet. The web was new, Netscape Navigator was the browser, and every website was hand-coded HTML with inline styles. There was no CSS (not until 1996), no JavaScript frameworks, no build tools. You wrote HTML in a text editor, uploaded it via FTP, and refreshed the browser.
That year, I built my first commercial website. The client was amazed that their business could be "on the Internet." I was amazed that someone would pay me to do this.
The Middle Years: Scale and Complexity
Over the next two decades, WebHouse built over 1,000 websites and 50+ mobile apps. We rode every wave:
- Late 90s: Table-based layouts, animated GIFs, Java applets
- Early 2000s: CSS, PHP, MySQL, the LAMP stack era. I built ODEUM, a custom CMS
- 2008: A detour to Dubai for IoT business development
- 2010s: WordPress dominance, responsive design, mobile-first thinking
- 2016-2026: A decade at Senti.Cloud, building an IoT platform while running WebHouse in parallel
Each era taught something. Table layouts taught me that constraints breed creativity. The CMS era taught me that content modeling is the hard problem. Mobile taught me that performance is a feature. IoT taught me that systems must be designed for failure.
> In 1996, I won a European Commission Award at the 5th International WWW Conference in Paris. The web was young, and so was I. Thirty years later, the excitement hasn't faded — it's just shifted from what's possible to what's elegant.
2025: The AI Pivot
The shift to AI-native development in 2025 was the most significant change since the web itself. Not because the tools are impressive — they are — but because they change the economics of software. What used to require a team of five now requires one developer with a well-structured project and an AI partner.
I don't say that to diminish teamwork. I say it because it means a solo developer in Blokhus, Denmark can build and maintain systems that would have been impossible a few years ago. That's the real revolution.
What Stays the Same
Through every technological shift, some things haven't changed:
- Clarity wins. Whether it's HTML in 1995 or TypeScript in 2025, the code that lasts is the code that's easy to read.
- Ship it, then iterate. Perfectionism is the enemy of progress. Get it live, get feedback, improve.
- The user doesn't care about your stack. They care if the thing works, loads fast, and solves their problem.
Thirty years in, I'm building the best software of my career. The tools have never been better, and the problems have never been more interesting.