I Gave a Three-Cent Coding Agent a Desk in the Homelab and Made Claude Its Reviewer
Claude Code is a great senior engineer and a terrible use of money for “add a test for row 144.” That one task doesn’t need judgment; it needs a person (or a cog) who will read the spec, write the test, run the suite, and report back honestly. Paying Opus-level tokens for that is like chartering a helicopter to cross the creek.
So I hired a cheaper junior. A DeepSeek-backed agent runs in a container on the homelab’s GPU box, paid per job instead of per month, briefed in writing, and — this is the whole trick — with Claude kept for the one part that actually needs a senior: reviewing its work. Claude reviews; I merge. About twenty jobs in, the agent’s bill has not dented a ten-dollar OpenRouter credit, and the process has taught me more about my own briefs than about the agent.
Moving 545 Passwords from Bitwarden's Cloud to Vaultwarden: the YubiKey Detour and the Count That Lied
For years I kept my password vault in Bitwarden’s cloud on a Premium subscription. It worked, it was cheap, and the clients were lovely. Then Bitwarden’s licensing direction this year gave me the nudge, on top of a reason I’d already settled on: I wanted the vault on hardware I own, behind my own TLS and my own backups.
The clients stay exactly the same. Vaultwarden speaks the Bitwarden API, so the browser extensions, desktop app and phone apps just get a new server URL — the change is where the bytes live, not how I touch them.
What follows is the honest version: a migration designed so my master password never passed through the AI assistant I do my homelab work with, a YubiKey detour of my own making, and three item counts for one vault that disagreed until I learned which to trust.
Speaker-Diarized Call Transcription on a CPU — No Cloud, No GPU, No Hugging Face Token
I wanted a speaker-labeled, timestamped transcript of a two-person phone call — the kind an LLM can actually reason about (“what did the customer commit to?”, “when did the price come up?”) — without shipping the audio to a cloud speech-to-text service.
One hard constraint shaped every decision: it all had to run on hardware already in the rack. No new GPU, no API keys, no “just use the cloud for the hard part.” A 35-watt desktop chip that already earns its keep running other things.
It worked. And, as usual, the interesting parts weren’t in the happy path — they were in why the whisper I already had running couldn’t do the job, how to diarize speakers on a CPU with no Hugging Face token, and a genuinely great debugging story that ends with a 114-byte file quietly corrupting every timestamp in the system.
A note on the transcript examples below: the real test call involved an actual third-party business and named individuals who never agreed to be quoted. Every name, company, and phone number in the excerpts here is synthetic — invented for illustration. In many US states, Colorado included, recording calls and reusing what was said carries real legal weight. If you build one of these, redact before you publish. I’m practicing what I preach.