GKGlyphKnitVin Rao
Work

Public project

Paperwake Document Archive

A portable document migration and archive stack using Paperless-ngx, Docker services, encrypted backup, and migration runbooks.

Docker ComposePaperless-ngxPostgresRedisCaddyBorgBackup

Context

Personal infrastructure project for moving document scanning and archiving away from Microsoft services.

Problem

Personal documents need searchable PDFs, exportable metadata, portable storage, and backup/restore confidence without depending on a single consumer cloud product.

Role

Designed the local and production Docker shape, migration documentation, backup/restore notes, and operational runbooks.

Constraints

  • Private documents and imports stay out of version control.
  • The archive must remain boring, portable, and recoverable.
  • ML additions must not make source-of-truth document records untrustworthy.

Approach

Used Paperless-ngx as the archive and metadata layer, Postgres and Redis for backing services, Caddy for production routing, and BorgBackup/Storage Box planning for encrypted offsite backup.

Challenges

The project balances migration automation with strict separation between safe repo documentation and private source files.

Impact

  • Operational evidence: documented local stack, production shape, migration runbook, backup planning, and private-source separation.
  • Repo evidence: architecture docs, backup/restore docs, migration runbook, local disposable stack, and production deployment notes.

Recruiter takeaway

Shows systems thinking beyond UI automation: infrastructure, data portability, privacy, and runbooks.

Engineering manager takeaway

The clean separation between source data and repo docs is exactly the kind of judgment production teams need.