Explorations and learnings from a collaborative re-design with Claude AI
My previous design for this website lasted over 12 years. Last week, I updated the design with Claude AI and now it's different, as is my perspective.

My decision to use AI was based on two factors.
1) I want to be in front of a computer less
I've lost interest in spending any more time in a front of a computer screen than what's required of me.
In years past, countless early mornings before work and late evenings after I could be found agonizing over pixel placement, line heights, and troubleshooting and fixing layout bugs. That's the effort that would have been required to research, create, build, test and refine a new design.
At this point in my life and career, I instead focus my time and energy on growing our business, supporting our team and improving our product. I'm far removed from daily in the trenches design work.
2) Yet, I still want to keep learning and progressing
AI is playing a much larger role in how many people build, ship, oversee and maintain digital products.
There's still humans needed to assess, refine and make decisions, but overall we're morphing from assemblers into conductors. Despite being in this industry for nearly 30 years, I still want to learn and experiment with new methods and approaches.
Thus, I spent a little time with Claude, and together we came up with what's here. And honestly, the collaborative experience was pretty fun.

I wanted to keep the same text and images, same structure and same URLs, but make a few key upgrades:
Dark theme, with light theme toggle
I had a dark themed personal website back in 2006 - 20 years ago - that was always one of my favourites (no screenshot, sorry). As such, I wanted a dark site but with the ability for people to easily toggle a light theme.
Bigger and more roomy
Speaking of design, my site's text and navigation links are now bigger and easier to tap on a phone. There's also more white space, increased contrast, new typography and thinner lines.
Overall, I think the design work section looks nicer (despite some small issues I still need to address) and the line length in longer entries, such as in the journal is shorter. Overall, I'd say my website is now easier to navigate and use.

No more CMS software
I don't want to deal with CMS software updates or complexities anymore.
The previous site was run on Kirby and this new version is just pure HTML and CSS with no JavaScript.
Given that over the past four years, I've not published more than four blog posts a year, I don't need a CMS or really, any extra technology layers. I'm comfortable hand-bombing HTML and updating the RSS feed on my own.
RSS feed fixed
Speaking of RSS, I finally fixed the RSS feed that's been broken for a couple of years. Oops?
The break first happened when I upgraded Kirby from an ancient edition to a newer one. I'd tried to troubleshoot and fix the RSS feed myself a few times, but after a while just gave up (see reason above).
Security improvements
I work in cybersecurity and the past few years have taught me a lot about what this means on a professional and personal level. Thus, I wanted my own website to meet higher security standards with a smaller attack surface, improved compliance and SEO, as well as increased trust, simply out of principle.

Accessibility improvements
I also wanted my website to meet higher accessibility standards. The previous version was WCAG 2.0 AA compliant, but the new version strives for WCAG 2.2 AAA.
This is important to me for many reasons, but it was also a personal challenge. What would it take to get there, stay there and what are the trade-offs? I sought to understand and grasp these concepts even further.

In summary
This is an interesting time to be in product design and development, but also the SaaS industry. There's so much turnover and experimentation, such rapid progress, that it all feels pretty wild, chaotic and to be honest, kind of scary.
Regardless, I've chosen to jump in because why not?
I can either choose to get left behind and stand on the dock watching others, or I can decide to dive headfirst and start swimming.