What this is
The Homelab is my operational infrastructure lab: the place where I run and rebuild virtualization, storage, networking, backup, self-hosted service, and local AI infrastructure work. It is the environment I operate, break, recover, document, and use to test infrastructure decisions before they become public claims.
The current lab includes Proxmox VE virtualization, Linux VMs and containers, TrueNAS-backed storage, dedicated Docker hosts, segmented networking at an abstract level, Proxmox Backup Server, and service operation across a mix of stable systems and active rebuild work.
Why it exists
As a kid it was game servers and Plex. As an adult it became personal services, agency, and learning. As a professional it is proof of concept, dev workflow testing, and something to point at when words need evidence.
The hardware changes constantly — Theseus’s ship, one component at a time — but the purpose stays the same: build it yourself, for yourself, and learn something every time it breaks.
Implementation layer
homelab-iac serves as the private implementation and recovery layer inside
the Homelab. I use it to turn specifications and operational decisions into
reviewed infrastructure artifacts while keeping the private repository,
inventories, access methods, and topology out of the public site.
The useful public story is the workflow rather than the inventory. The repository holds OpenTofu declarations and virtualization modules, Ansible roles and playbooks, a Packer image/template lane, sanitized service reconstruction bundles, validators and tests, ADRs, runbooks, status records, command-packet workflows, and closeout evidence. Those artifacts support repeatable change, but live infrastructure mutation remains human-governed.
Current evidence
The lab currently supports operating, troubleshooting, documenting, and iterating on self-hosted infrastructure. The verified evidence I can summarize publicly includes source-control recovery, protected Git behavior, application restore work, mirror validation, and reviewed VM retirement through OpenTofu.
I treat tracked declarations as source evidence first, then validate and promote the parts that are ready for live change. The durable pattern is moving from specification to reviewed source, then from reviewed source to validation, recovery notes, and human-approved infrastructure changes.
Boundaries
I keep the topology abstract on purpose. Hostnames, addresses, VLAN identifiers, ports, storage exports, backup identifiers, device maps, private repository remotes, command text, and live-state details stay private.
The lab includes validated experiments and rebuild work alongside currently operated services. Remote state, unattended deployment, complete CI/CD, active proposal automation, self-healing reconciliation, expanded offsite recovery, and broader automation maturity remain incomplete or planned unless I describe them separately as validated.
Next work
The next layer is tightening the rebuild path: carrying approved specifications
into homelab-iac, validating recovery and service reconstruction flows,
improving backup and restore discipline, and continuing to separate live
authority from proposal and validation workflows.
Phred remains independently evidenced AI infrastructure work. The Homelab is the broader operating environment around it, and future Phred rebuild work will need its own validation before I present it as managed infrastructure.