It’s been a while.
Actually, “a while” is a severe understatement there. It’s been years. The last time I wrote anything about Android on my personal blog, Kotlin was still considered exotic by some people, Compose was still wishful thinking and RecyclerView was a fresh wound on everyone’s soul. Good times.
I won’t bore you with the usual “life got busy” speech. You know how it goes. The blog quietly rotted, the old domain expired (now home to what I hope is a successful and honest Chinese business, instead of a gambling scam like Gemini suggested), the CI pipeline that was held together with duct tape and good intentions finally gave up. I did not intervene. We parted ways peacefully.
So why now?
Honestly? AI. LLMs. The whole thing.
I’ve been doing Android for long enough to remember when the hot debate was whether Fragments were a good idea. I’ve lived through the Java-to-Kotlin migration, the architectural pattern wars, the endless recycling of “this new library will solve state management forever.” I guess I got comfortable. Things were good, but nothing was new new.
Then, large language models showed up. I really didn’t care at first, or not for producing software at least: ChatGPT was cool to plan trips and ask random stuff, and my mum has been very happy ever since that I taught her how to use it to ask about what to cook for dinner or how to tend to her plants.
Time passed, and the LLMs evolved. They got better. Like, a lot better. To the point where I couldn’t ignore them anymore. So I eventually started playing around with them. By the time I noticed, I was already three different apps in (wonder if I’ll ever finish those), going to bed at ridiculous hours because I didn’t notice time passing. In a completely unexpected way, there it was. That feeling again — the one I had years ago when I first started tinkering with Android and every week felt like someone had unlocked a new room. That slightly uncomfortable excitement where you don’t fully know what you’re doing yet but you can’t stop thinking about it.
Even more unexpected was where it pointed: straight back at Android. That spark I thought had just quietly settled was back, and with it came something I hadn’t felt in a while — the genuine urge to learn more, to be active in the community again, and to contribute. To write about the platform I’ve spent years on, and hopefully put something useful back into a community that’s given me a lot.
So here we are. The blog is back, rebuilt from scratch (Astro this time, no more Jekyll held together with prayers). As you might expect, the rebuild was AI-assisted — I had Claude set the whole thing up, and it took a few hours instead of the usual “I’ll get to it eventually” three years.
As for what I’ll write about: mostly Android. That’s still where I spend most of my time, and there’s always something worth digging into — architecture, Kotlin quirks, the occasional hill I’m willing to die on. Whether AI ends up woven into those posts or not, I genuinely don’t know yet. Sometimes it’ll be relevant. Sometimes it won’t. I’m not going to force it either way.
You might be asking yourself something like “writing? On the age of AI where all the information I need is a prompt away?” Well, yeah. The answers for your prompts have to come from somewhere, provided the agent doesn’t hallucinate. The internet is increasingly full of AI-generated slop, and the models that come after the current ones are going to be trained on whatever we leave behind. I’d like to think that people who actually know what they’re talking about should probably be the ones putting stuff out there — instead of leaving the field wide open for confidently-written nonsense. So consider this my small act of responsible AI stewardship: feeding the future machines with quality content. Or at least content I think is quality, which is not quite the same thing, but let’s not dwell on that.
Anyway, see you on my next post. No promises on frequency. But the buzz is back, and that’s usually enough.