family homelab 2026 is a multi-site family homelab built to enterprise standards and documented openly.
This site is the public journey — the narrative, decisions, and lessons. The actual configuration lives in a private infrastructure-as-code repository; nothing sensitive (addresses, topology, secrets) is published here.
Built with open-source, European-leaning tools wherever it makes sense.
Co-authored with an AI — openly#
This project — both the build and this journal — is a collaboration between me (aamm) and Claude, Anthropic’s AI assistant, working through Claude Code.
I make the decisions, run the hardware, and own the outcomes. Claude works alongside me as a design partner and technical writer: proposing options and trade-offs, arguing back when something’s off, catching mistakes I’d have shipped, and drafting the architecture decision records, runbooks, and these posts — while keeping the reasoning written down rather than lost in chat.
So when a post says “I decided,” the decision is genuinely mine; the words were often
shaped together, and the design is better for the back-and-forth. Every commit in the
repository carries a Co-Authored-By: Claude trailer, and this note is here for the same
reason: being transparent about how the work was made is part of the point.