dads2busy.github.io

Personal academic portfolio website + curriculum vitae for Aaron D. Schroeder, Ph.D. — Research Associate Professor, Social & Decision Analytics Division, University of Virginia Biocomplexity Institute.

🔗 Live site: https://dads2busy.github.io 📄 Vita PDF: https://dads2busy.github.io/vita.pdf

Architecture in 60 seconds

One YAML file is the single source of truth. Tooling renders it into multiple outputs.

                          Data-build layer (write to SSOT)
                          ┌───────────────────────────────────┐
External sources          │ scripts/analyze_sources.py        │ ← LLM-extracts CV/DOCX
  CV / DOCX     ─────────▶│ scripts/orcid_works.py            │ ← ORCID API + diff
  ORCID API               │ scripts/fill_missing_fields.py    │ ← propose blanks to fill
                          │   (all emit review diffs;          │
                          │    NEVER auto-edit profile.yaml)   │
                          └───────────────────────────────────┘
                                          │ human paste
                                          ▼
                         ┌───────────────────────────────────┐
                         │  site/content/profile.yaml (SSOT) │
                         │  site/content/site_extras.yaml    │
                         └───────────────────────────────────┘
                                          │
                          Delivery layer (read from SSOT)
                          ┌───────────────────────────────────┐
                          │ scripts/profile_to_content_jsons  │
                          │   → 5 site content JSONs           │
                          │   → website pages render            │
                          │                                    │
                          │ rendercv render → vita.pdf         │
                          │                                    │
                          │ site/src/app/page.tsx              │
                          │   → home page (reads YAML directly)│
                          └───────────────────────────────────┘

Full architecture details: see ARCHITECTURE.md.

Where to make common changes

Add or edit a publication

Edit site/content/profile.yaml — find the matching publication-style section (Refereed Journal Articles, Book Chapters, Research / Technical Reports, etc.) and add/edit the entry. Use the existing entries' shape as a template.

Add or edit a presentation/talk

Edit site/content/profile.yaml — Presentations: section. Set subcategory: to one of: Panelist, Presentations/Workshops, Committee, Lecture, Expert Forum, Expert Webinar.

Add or edit a research project

Edit site/content/profile.yaml — Research Projects: section. Fields: name, funder, award, role, date, summary, slug.

Update bio / education / skills / awards (home page)

Edit site/content/profile.yaml — Summary, Education, Skills, Awards & Honors sections. The home page reads these directly via js-yaml at build time.

Change vita PDF look

Edit design.theme: in profile.yaml (currently classic). RenderCV ships several built-in themes; full customization via custom RenderCV themes.

Bulk-import new entries from a CV/DOCX

Drop the file into sources/cvs/, then run:

.venv/bin/python scripts/analyze_sources.py

Review sources/_diff.md (gitignored) and hand-paste accepted entries into profile.yaml. The script never touches profile.yaml directly.

Sync new publications from ORCID

.venv/bin/python orcid_works.py

Review orcid_diff.md (gitignored). Same paste workflow.

Sync new datasets from Zenodo

.venv/bin/python scripts/zenodo_works.py

Review zenodo_diff.md (gitignored). Same paste workflow — proposed entries land in profile.yaml's Data & Software section.

Find missing fields across profile.yaml

.venv/bin/python scripts/fill_missing_fields.py

Reviews profile.yaml for blank name/category fields, suggests derivations, writes profile_fill_diff.md.

Local development

Setup (one-time)

# Python venv for build tooling + RenderCV
python3 -m venv .venv
.venv/bin/pip install -r scripts/requirements.txt

# Node deps for the website
cd site && npm ci

Set up .env at the repo root with:

ORCID_CLIENT_ID=APP-...           # for orcid_works.py
ORCID_CLIENT_SECRET=...
ORCID_ID=0000-0003-4372-2241
ANTHROPIC_API_KEY=sk-ant-...      # for analyze_sources.py

Run the dev server

cd site && npm run dev   # http://localhost:3000

The predev npm script automatically regenerates the 5 content JSONs from profile.yaml before starting the dev server, so any YAML change is visible after the next page reload.

Build the static site locally

cd site && npm run build

Output lands in site/out/. The prebuild npm script regenerates JSONs first.

Generate the vita PDF locally

cd site && npm run gen:vita

Output: site/public/vita.pdf (gitignored). CI also runs RenderCV on every push.

Run tests

.venv/bin/pytest scripts/tests/

108 tests cover the data-build/delivery libraries.

Tech stack

Layer	Tools
Frontend	Next.js 16 (App Router, TypeScript, static export), Tailwind CSS v4, Google Fonts (Inter / Instrument Serif / Geist Mono)
YAML loading	`js-yaml` + `zod` (validates `site_extras.yaml`) at build time
SSOT	`site/content/profile.yaml` (RenderCV-format YAML with custom keys)
Vita PDF	RenderCV (Python) — `classic` theme
CV / source analyzer	Anthropic Claude API (`claude-sonnet-4-6`) + `mammoth` for DOCX
ORCID sync	Python `urllib` against ORCID Public API v3.0
Tests	pytest (Python side); Next.js build is the implicit test on the JS side
Deployment	GitHub Actions → GitHub Pages

Repository layout

.
├── site/                          # Next.js project
│   ├── content/
│   │   ├── profile.yaml           # SSOT — edit this for content
│   │   └── site_extras.yaml       # Site-only display fields
│   ├── src/
│   │   ├── app/                   # App Router pages
│   │   ├── components/            # WritingCitation, SpeakingCitation, etc.
│   │   └── lib/                   # types, content loader, schemas
│   └── public/                    # Static assets, generated vita.pdf
├── scripts/
│   ├── profile_to_content_jsons.py    # SSOT → 5 site JSONs (delivery)
│   ├── json_emitters.py               # Per-section converters
│   ├── analyze_sources.py             # Plan G: CV/DOCX → review diff
│   ├── analyzer_lib.py                # Source analyzer helpers
│   ├── orcid_diff_lib.py              # Plan E: ORCID diff helpers
│   ├── fill_missing_fields.py         # Plan H: blank-field review tool
│   ├── profile_derive_lib.py          # Plan H: title/category derivation
│   └── tests/                         # pytest suite (108 tests)
├── sources/                       # Drop-zone for raw artifacts
│   ├── cvs/                       # Old vitae (DOCX) for analyzer ingest
│   └── README.md                  # Drop-zone workflow
├── orcid_works.py                 # Plan E: ORCID fetcher + diff
├── orcid_works.json               # ORCID snapshot (committed)
├── docs/superpowers/plans/        # Implementation plans (Plans A–H)
├── ARCHITECTURE.md                # Data-build vs delivery layer details
├── SSOT_PLAN.md                   # Higher-level SSOT vision
├── .github/workflows/deploy.yml   # CI: render vita + build + deploy
└── old jekyll site/               # Archived original Jekyll site

Contributing / extending

Add a new delivery output (e.g. CSL JSON for Zotero, LinkedIn paste view): write a pure Python emitter in scripts/, mirror the profile_to_content_jsons.py pattern, add shape-completeness tests.
Add a new section to profile.yaml: append to cv.sections.<NewSection>. RenderCV will pick it up automatically. The website's content pages only render sections explicitly mapped in WRITING_SUBCATEGORY (in profile_to_content_jsons.py).
Add a new delivery template: each delivery has its own format/template — RenderCV theme for the vita, React components for the website. Don't push delivery-specific formatting back into profile.yaml.

See ARCHITECTURE.md for the design principles, and docs/superpowers/plans/ for the historical implementation plans (each documents a chunk of the migration from the original Jekyll site).

License

Content is licensed under Creative Commons Attribution 4.0. Code is MIT-licensed where not otherwise noted.

Navigation Menu

Repository files navigation

dads2busy.github.io

Architecture in 60 seconds

Where to make common changes

Add or edit a publication

Add or edit a presentation/talk

Add or edit a research project

Update bio / education / skills / awards (home page)

Update tagline or polymath callout

Change presentation category order on `/speaking`

Change citation rendering

Change vita PDF look

Bulk-import new entries from a CV/DOCX

Sync new publications from ORCID

Sync new datasets from Zenodo

Find missing fields across profile.yaml

Local development

Setup (one-time)

Run the dev server

Build the static site locally

Generate the vita PDF locally

Run tests

Tech stack

Repository layout

Contributing / extending

License

About

Releases

Packages

Contributors

Languages

Name		Name	Last commit message	Last commit date
Latest commit History 716 Commits
.github/workflows		.github/workflows
advisor		advisor
docs/superpowers		docs/superpowers
old jekyll site		old jekyll site
scripts		scripts
site		site
sources		sources
.gitignore		.gitignore
ARCHITECTURE.md		ARCHITECTURE.md
LICENSE		LICENSE
README.md		README.md
SSOT_PLAN.md		SSOT_PLAN.md
orcid_works.json		orcid_works.json
orcid_works.py		orcid_works.py

Folders and files

Latest commit

History

Repository files navigation

dads2busy.github.io

Architecture in 60 seconds

Where to make common changes

Add or edit a publication

Add or edit a presentation/talk

Add or edit a research project

Update bio / education / skills / awards (home page)

Update tagline or polymath callout

Change presentation category order on /speaking

Change citation rendering

Change vita PDF look

Bulk-import new entries from a CV/DOCX

Sync new publications from ORCID

Sync new datasets from Zenodo

Find missing fields across profile.yaml

Local development

Setup (one-time)

Run the dev server

Build the static site locally

Generate the vita PDF locally

Run tests

Tech stack

Repository layout

Contributing / extending

License

About

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages

Change presentation category order on `/speaking`