Follow along as I build a production-ready Kubernetes cluster on Raspberry Pi 5s.
I’m documenting everything: wins, failures, and lessons learned while working with AI.
Part 2: Services, Services, Everywhere!
Part 2 - Services, Services, Everywhere Working on the configurations in Part 2 has been an incredibly frustrating, but incredibly rewarding experience. From the first milestone of getting the hardware built and core services running, I’ve been quickly adding to that foundation. In some cases, even more than what I had expected. At times, the configuration challenges seemed insurmountable. But time and again, I was able to troubleshoot with Claude and in some cases end up even better than the original plan. ...
Blogging with Google Gemini
The Two Brains: How AI Refined My Home Lab Narrative Building a Kubernetes cluster on a stack of Raspberry Pi 5s is the technical grunt work. But turning my build into a readable blog series? That’s where the real creative pain starts. Turns out, I could continue my AI experiment, which is how my content planning partnership with Gemini took off. I had Claude locked down as the technical expert—generating technical details and expediting my build progress. I trust Claude to help me knows which flags, commands, and configs to use. But I enlisted Gemini to act as my content planning partner—the one who could make sense of the mess for a reader. It’s a classic division of labor: Claude for the execution, Gemini for the narrative. And frankly, I feel it has helped me consolidated my thoughts, notes, and approach into something readable. ...
Part 1: Core Cluster Hardware Build and Initial Install
Part 1 - Cluster Build As I mentioned in this series’ first post —this started as an abstract idea and I’ve been working with AI to refine and get to a point where I am physically implementing this home lab project. Spoiler alert—the core cluster is live! No public services yet, but K3s is running and my foundational configuration is set. ...
The Cluster Chicken & Egg: How to Blog About a Server That Doesn't Exist
My entire goal for this project is to self-host everything, including the blog you’re reading right now, on a Raspberry Pi 5 K3s cluster. That presents a small, hilariously meta problem: How do I blog about building the server when the server hosting the blog doesn’t exist? The Interim Solution: Hugo + Cloudflare I turned to Claude for a recommendation on a temporary, down-and-dirty setup that I could spin up instantly. I wanted to be able to launch in a day, spend no money, and be able to migrate easily. ...
The Spark - Why I'm Building a Raspberry Pi 5 Cluster for AI & Blogging
You know how it starts? A simple thought that doesn’t go away. So I started a Claude Project, and issued a prompt: could I actually build a tiny, personal data center on my desk using just a few Raspberry Pi 5s? My main goal was pretty ambitious: I wanted to make a mini-cloud. Think of it as replicating basic services like AWS EC2, DynamoDB, and Lambda, but with open-source tools instead. This way, I could host my own Python and Node.js apps, run a few small websites, and set up a proper sandbox where I could genuinely collaborate with AI models like Claude and Gemini. ...