Project Holocron
activeProject Holocron enables easy updated to my resume and website by keeping a single source of truth document
Computer ScienceWeb DevServerhostingPython
Overview
This site, and the resume it generates. Project Holocron keeps a single source of truth for my professional history, publications, and projects, then compiles that one dataset into two very different outputs: a typeset PDF resume and a static portfolio website. Updating a job, a paper, or a project means editing one YAML file — never reconciling a document against a webpage.
Technical Details
- Structured all professional data as a central
vault.yaml, with projects and publications as markdown files carrying typed frontmatter - Built the resume engine in Python, injecting vault data into a LaTeX template through Jinja2 with custom delimiters so Jinja’s syntax doesn’t collide with LaTeX’s braces
- Built the site in Astro, reading the same vault directly and validating every markdown file against a schema at build time, so a malformed entry fails the build instead of silently rendering wrong
- Automated the whole pipeline in GitHub Actions: each push to
mainrecompiles the PDF, rebuilds the site, and deploys to GitHub Pages