How this site works
This site is a wiki built in Quartz, maintained as I go. It runs as a working record of the practice, projects, decisions, playbooks, things I’ve read, rather than a résumé refreshed for applications.
Two-stage build
The repo has two layers:
raw/, source material. Defuddled web captures, per-project evidence dives (I read the code and extract architecture, decisions, metrics), PDF extracts, pasted notes. Re-runnable. Not edited by hand.content/, the wiki you’re reading. Written by hand from the raw layer, never auto-stitched from it. Keeping them separate keeps the written layer in my voice and the raw layer re-runnable.
Folders
skills/, what I do, one page per speciality, with the projects that back it.projects/, one page per build, with the decisions and outcomes attached.decisions/, positions I hold, with reasoning.playbooks/, repeatable recipes I want to run the same way twice.notes/, short-form dated thinking.influences/, video, audio, and writing that shaped the work.open-questions/, things I haven’t resolved.
Dated entries
Pages carry dates where it matters. If a position shifts, the old page stays and a new one lands next to it rather than overwriting. Both are readable from the Decisions folder.
Plain markdown in git
The whole site is markdown files in a public git repo, built with Quartz and hosted on Vercel. No CMS, no database, no platform lock-in. Works as a local Obsidian vault too, that’s how I maintain it.
Why this shape
A CV compresses the work into a page. Keeping the detail, decisions, playbooks, the reasoning behind them, lets readers follow the thread they care about, and lets the site grow with the work.
