Skip to content

Fix populate_db notebook for production seeding - #90

Merged
androemeda merged 1 commit into
mainfrom
fix/populate-db-notebook
Aug 9, 2026
Merged

Fix populate_db notebook for production seeding#90
androemeda merged 1 commit into
mainfrom
fix/populate-db-notebook

Conversation

@androemeda

Copy link
Copy Markdown
Collaborator

Reworks backend/populate_db.ipynb to seed the deployed Qdrant KB end-to-end.

Changes:

  • Point uploads at the deployed backend (vcell-ai-dev.cam.uchc.edu/api) instead of localhost
  • Load admin JWT from .env and attach as Bearer header on every request
  • Delete-before-upload per filename for idempotent re-runs (backend has no dedup)
  • Pass source_url on each upload so origin is preserved in the Qdrant payload
  • Clean up noisy prints and stray Colab-isms

Seeded the dev KB with 153 files across the three configured source directories.

@androemeda
androemeda merged commit 5b350dc into main Aug 9, 2026
1 check passed
@androemeda
androemeda deleted the fix/populate-db-notebook branch August 9, 2026 20:30
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant