Scanning pages with the mobile app
The mobile scanner uses Google's native ML Kit document scanner (Android) or Apple's VisionKit (iOS) — the same one Google Drive and Apple Notes use. We add durable batch storage and one-tap upload on top.
Flow
- Tap New scan on the Dashboard or a project.
- Choose a mode: Single (default), Side-by-side, or
Top / bottom. See "Scanning IDs and two-up mode" for the 2-up options.
- Tap Start scan. The native scanner takes over — capture as
many pages as you want, then tap Save in the scanner.
- The app shows "Saving page N of M…" while it prepares each scan
one at a time and copies it into durable storage.
- On the Review batch screen, remove any bad captures, pick a
target project, and tap Upload.
Tips
- Scan on a flat surface with even light. The scanner auto-crops and
runs a mild B&W filter, but it can't rescue a shadow across the page.
- Aim for the whole page inside the viewfinder — cropping outside
the frame is easier than fixing under-scanned edges later.
- Fine text on technical drawings needs a steady hand. If the scan
is blurry, retake it before uploading — replacing a page's source image after upload is possible but re-runs the whole pipeline.
Why "Saving page N of M…" instead of instant
Processing scans one at a time is deliberate: 100 captures processed in parallel used to spike memory and get the app killed by the OS, losing every scan in the session. Sequential means each page lands on disk before the next one starts — if the app dies mid-batch you resume from where you stopped, not from zero.