How Tailoring Works

The Mental Model

Think of resume.md as a database of everything you've done. Each application is a query against that database.

resume.md       (all your career data)
  × tags        (WHERE — filter what's relevant)
  × hide/pin    (SELECT — which sections, in what order)
  × vars        (SET — role-specific keywords, taglines)
  × style       (FORMAT — presentation adjustments)
  → output      (PDF, HTML, DOCX, or PNG)

Your career data is durable, it accumulates over years. The way you present it changes with every application. Tags filter the content, views shape the output.

Tags and Views

Two independent concerns combine to produce each render:

	Purpose	SQL analogy
Tags	Filter content	WHERE
Views	Configure rendering	SELECT + FORMAT

Tags decide what's in and what's out. Add {.@frontend} to a bullet and it only appears when rendering for frontend. Compose tags (fullstack: [frontend, backend]) to build reusable content sets. Tags are purely filters, they never touch rendering.

Views decide how filtered content renders: which sections appear, in what order, what keywords to inject, how many pages to target. Every render uses a view, even if you never create one explicitly.

Both are optional and independent. You can use tags without views, views without tags, or both together. A resume with neither produces a single PDF with everything included.