Content Updates

This site now has a single content sync pipeline so the timeline, stream, personal page, and related generated data stay aligned.

Source of truth

Posts: add/edit files in _posts/
Career, education, projects: edit _data/career.yml
Publications: edit publications/index.markdown
Manual social items: edit _data/stream_manual.yml
Manual photos/media: edit _data/photos_manual.yml
YouTube videos: fetched into _data/videos.yml

Do not manually edit these generated files unless you have a very specific reason:

_data/stream.json
_data/photos.json
_data/timeline.json
_data/videos.yml for normal YouTube refreshes

Update locally

Fetch external sources and rebuild everything:

python3 scripts/sync_site_data.py

Rebuild only derived files from the current local data:

python3 scripts/sync_site_data.py --skip-fetch

What gets rebuilt

scripts/fetch_stream.py builds _data/stream.json from Medium, local posts, publications metadata, and manual social entries.
scripts/fetch_photos.py builds _data/photos.json from Flickr and manual media entries.
scripts/fetch_videos.py syncs YouTube metadata into _data/videos.yml.
scripts/build_timeline.py builds _data/timeline.json from career, publications, photos, videos, and stream data.

GitHub Actions

.github/workflows/sync_site_data.yml runs on a schedule and can be triggered manually. It syncs generated data, validates a Jekyll build, and commits generated files if they changed.
.github/workflows/build_site.yml runs on pushes and pull requests. It rebuilds derived data from local sources and checks that the site still builds.

Typical update flow

Edit the source files you actually own, such as _posts/, _data/career.yml, publications/index.markdown, or the manual data YAML files.
Run python3 scripts/sync_site_data.py --skip-fetch if you only changed local source files.
Run python3 scripts/sync_site_data.py if you want fresh external data too.
Review the generated diffs, especially _data/timeline.json.
Commit the source changes and the regenerated data together.