Hello internet! After a brief hiatus from online posts, I once again implore you to suffer through my ramblings, albeit this time hopefully a little more focused. Wait — it’s been 20 years since the Xanga / GeoCities days?!?

Apparently so. Which means I’ve officially reached the age where the internet I grew up with feels like an abandoned mall: still standing, slightly dusty, and weirdly comforting.

A blog?!? In this economy!?

Well I have the luxury of hosting this on Cloudflare for free so at least there’s that

It’s boggling to me how much things have changed since my last opportunity to post. It used to be that I was wowed by WYSIWYG editors and trying to hack custom CSS on a hosted blog. Now I not only get to but choose to write in markdown. #nerd

Starting a (primarily tech) blog in 2026 feels almost contrarian. We’re surrounded by tools that can generate endless text on demand, feeds that reward immediacy over thought, and platforms that quietly discourage anything longer than a screenful. Every few months I’d think, I should really write some of this stuff down, and then immediately talk myself out of it because there are already too many words on the internet. So why add another blog post to the pile?

Because writing things down still matters. Mostly technical things — patterns I’ve noticed, approaches that worked, mistakes I’d rather not repeat. The kind of stuff that tends to live in my head, or in a graveyard of private notes, until it quietly disappears.

This site isn’t meant to be optimized, branded, or particularly impressive. It’s a place to leave ideas in a form that survives longer than a scrolling thumb. No algorithm is waiting to rescue these words. If you’re reading this, you chose to be here, and I like that contract a lot.

Well technically

So many of the techniques and mental models I rely on came from blog posts written by people who took the time to explain how something worked for them. Mind you: many of these are not tutorials polished to perfection, just honest write-ups of “here’s the problem, here’s what I tried, here’s what stuck.” I spend entirely too much of my day browsing HackerNews for “this one simple trick”. It’s how I learn.

That’s the lane I’m aiming for here. In a lot of ways this is my attempt to give back.

If I write about a pattern, it’s because it helped me. If I share an idea, it’s because I tripped over it enough times that it finally lodged itself in my brain. If it saves someone else a bit of time or frustration, great. If not, at least I’ll have something to point to next time I forget why I made a decision.

On AI, and why it’s not the point here

I’d be lying if I said AI tools aren’t part of my workflow at all. They exist, they’re useful, and pretending otherwise would be silly.

But this blog is intentionally not an experiment in AI-generated writing. This is intentionally human-written, even when that means it’s a little uneven or longer than it needs to be. I’d rather have a slightly messy explanation that reflects real understanding than a perfectly smooth one that doesn’t.

If an LLM shows up here, it’s as a spellchecker, a sounding board, or reviewing for tone/message — it is not a ghostwriter. The responsibility for the ideas, the mistakes, and the inevitable bad takes is mine, end to end. That ownership matters to me.

Except for images — without some help, you’d be getting MS Paint–level diagrams.

And ironically, I’ve gotten better at using the em dash since it started appearing everywhere.

Meta posts as a survival strategy

This entry exists mostly so I can hit the proverbial “publish” button without overthinking into oblivion. First posts are weird. Once that hurdle’s cleared, the rest should be easier. Future posts will be more concrete: code, diagrams, tradeoffs, and the occasional “well, that didn’t work” retrospective. Some will age poorly. I’m okay with that. Stale ideas are proof that newer ones happened later.

For now, we can tick that “write first post” mental hurdle.

Thanks for hanging through the end of the first one. More soon — written slowly, imperfectly, and very much by a human.