Not Another README
If you’ve found your way here from one of my GitHub repos, you might be wondering: why a blog? The README was long enough already.
Here’s the thing: READMEs tell you what a tool does. They don’t tell you why it exists. They don’t capture the 2 AM debugging sessions where I realised the problem wasn’t the code, it was the approach. They don’t show you the gaps I was trying to fill, or what I learned when I filled them.
This blog is where that thinking lives.
The Gap Everyone’s Missing
There’s a misconception about AI right now. People think it does everything for you. The marketing says it, the demos show it, the influencers promise it.
The reality? AI does everything for you if you can fill the gaps.
And there are gaps everywhere.
I spend my time finding those gaps. Drilling into them. Testing whether solutions are real or just window dressing. Working out what’s actually missing versus what’s just poorly documented.
That’s what this blog is about: the gaps, the workarounds, the things I’ve learned from actually building with AI rather than just talking about it.
Someone in the Trenches
I’m not writing from a research lab or a conference stage. I’m writing from the trenches.
When I prototype something, I push it until it breaks. When I find a solution that works, I use it for real. When something claims to be revolutionary, I want to know if it actually delivers or if it’s theatre.
The tools I build come from real frustration with real problems. The Memory Bank repo that hit almost 3,000 stars? That came from watching AI forget my entire codebase between sessions. The Claude Code Collective? That came from wanting to see agents actually collaborate, pushing Claude Code as far as it would go before sub-agents even existed. When Anthropic released sub-agents, I pushed those too with a new release.
Each post on this blog will have a similar origin story: a gap I found, what I tried, what worked, what didn’t.
Why Long-Form Still Matters
We’re in an era of tweets and threads and short-form everything. So why bother with long-form blog posts?
Two reasons.
First, original thinking doesn’t fit in 280 characters. The nuance matters. The “here’s what I thought, here’s why I was wrong, here’s what I learned” takes space. Regurgitated content is everywhere. Original perspectives are rare.
Second, AI agents are reading the web. They’re training on content. They’re patrolling for context. If I want my thinking to be understood, whether by humans or by the models they’ll use to find information, I need to lay it out properly.
This isn’t just content. It’s a source of truth.
Own Your Source
I could write threads on X. I could post on LinkedIn. I could drop takes on Reddit.
But then I don’t own it. I can’t control it. It gets lost in algorithms and feeds.
This blog is my source. Everything syndicates from here. If I post something on LinkedIn or X, it started here first. My thinking, my reasoning, my perspective: all in one place, under my control.
That’s not about ego. It’s about having a foundation.
What to Expect
I’ll be writing about:
- The gaps in AI tooling: What’s missing, what’s overhyped, what actually works
- Context Engineering: The discipline that evolved from prompt engineering, now encompassing skills, agents, memory systems, and whatever comes next
- Tools I’ve built: The why behind the what
- Real-world patterns: From enterprise experience to open source experiments
- What I got wrong: Because learning happens in the failures
No marketing speak. No “delving into the ever-evolving landscape.” Just what I’ve found, what I’ve built, and what I think.
Find Me Elsewhere
- GitHub: vanzan01 (where the tools live)
If something here resonates, or if you’re working on similar problems, reach out. I’m always happy to chat.
Published: January 7, 2026 Author: Nick (vanzan01) Location: Perth, Australia