Homelab
13 posts
Breaking up with Claude Code
OpenCode with open models isn't a one-to-one swap for Claude Code. What broke, what fixed it, and which models I use.
My Development Setup in 2026
My development setup in 2026: Ghostty, plain zsh, Handy voice input, git worktrees for concurrent Claude Code agents, and chezmoi. What I switched and why.
I Made Adult Swim Bumpers for My Plex Server
I grew up on Toonami and Adult Swim, so I built a little CLI that makes Adult Swim-style bumpers and now every movie on my Plex server opens with one.
I Finally Stopped Managing My Homelab by Hand (Everything Just Kept Breaking)
How I migrated 62 Proxmox containers to OpenTofu with a self-hosted MinIO state backend - the audit, the wave migration, the stack, and whether immutable infrastructure is actually worth it.
How I Share Two Used GPUs Across Seven Services in My Proxmox Homelab
Proxmox GPU passthrough for LXC: how I share an 8GB Quadro RTX 4000 and a 16GB RTX A4000 across seven services, with local LLM speed and accuracy tests.
I Built a Fully Local Voice Assistant for Home Assistant (With GPU, No Cloud Required)
How I built a GPU-accelerated local voice pipeline for Home Assistant using faster-whisper, Piper TTS, and Ollama - zero cloud required, and the failures that got me here.
Self-Hosted Music Still Sucks in 2026
The *arr ecosystem perfected video automation but music remains stuck with album-centric workflows. Current tools like Lidarr force complete album downloads when users want individual tracks.
My Homelab Two Years Later, Every Upgrade and What It Cost
Two years after a $200 ThinkServer: two Dell R730s, 73 containers, metered power data, and whether a homelab pays for itself.
Implementing MikroTik's Binary API Protocol in Python from Scratch
A deep dive into implementing MikroTik's proprietary RouterOS binary API protocol in Python - variable-length encoding, sentence-based messaging, and programmatic network infrastructure control. Zero dependencies, 137 lines.
Building a GPU-Accelerated Subtitle Generator for My Video Library
How I built a self-hosted subtitle generation pipeline using GPU-accelerated Whisper AI on my homelab - with parallel workers, VRAM-aware scheduling, automatic language detection, and a multi-layer hallucination filter. A practical faster-whisper guide for video libraries at scale.
I Replaced Alexa with a Dumber Voice Assistant (But at Least It’s Private)
After five years of building automations in Home Assistant, I finally pulled the trigger on something I’d been contemplating for months: completely replacing my Amazon Alexa ecosystem with Home...
Adding Custom Domain Names on Your Internal Network Using Nginx Proxy Manager and Pi-Hole
How I set up custom domain names on my internal network using Pi-Hole for DNS and Nginx Proxy Manager for traffic control and SSL - including what actually works for SSL on local-only domains.
How to get started building a Homelab server in 2026
I built my first Homelab server for under $200 using a used Lenovo ThinkServer from Facebook Marketplace and Proxmox. Here is everything I learned about picking hardware, choosing an OS, setting up containers, and mounting NAS storage.