Private on-device OCR / SELECTED ENGINEERING WORK
MedList
Turn medicine-order photos and screenshots into editable lists on Android and iOS, entirely on device.
This overview describes my engineering work on this project, including its purpose, implementation, and technical focus. Project illustrations are abstract; private assets and internal implementation details are not published here.
Overview
Turn medicine-order photos and screenshots into editable lists on Android and iOS, entirely on device. My work connects the user experience with the data, application logic, and integrations behind these workflows.
Problem addressed
Medicine-order images need to become structured, editable records without sending them to a remote OCR service.
Solution
Built a Flutter app using PP-OCRv5 and ONNX Runtime for local text detection, recognition, and parsing, with review before saving.
Key features
- Photo and screenshot input
- On-device text detection
- Text recognition
- Confidence scoring
- Medicine-field parsing
- Editable lists
- SQLite audit storage
- Mandatory human review
Technical outcome
A private Android and iOS OCR workflow with no OCR API or runtime network dependency, plus local audit records and human review.
Interested in the engineering behind this work? Get in touch to discuss the approach and a similar product.