Inspiration
With traditional travel and expense programs, processing a single report can take an average of 20 minutes! In a recent study, we found that users can save an average of 3 hours per week on expense management. That's an additional 156 hours per year that individuals could put towards things that truly matter.
What it does
Receipify serves as an all-in-one platform for easily compiling your expenses. Users can simply take a picture of their receipts. Our machine-learning-based platform will employ OCR and NLP to interpret, categorize, and convert various components of the receipt to compile a PDF or CSV report neatly.
How we built it
Receipify is built on a react.js front end with a node.js backend. We employ the Veryfi OCR API and NLP model to sort and interpret various components of a receipt, including total cost, currency type, contact details, and expense type.
Challenges we ran into
We encountered problems with flask, as we initially planned to code it in python but later ran into many issues. Ultimately, we were forced to switch to node and code in js. Another issue was integrating our front end.
Accomplishments that we're proud of
In combination with an intuitive and modern UI, Receipify was consistently able to quickly and accurately process receipts from various countries with different formats.
What we learned
Despite not using it, we learned how to work with Flask. We also learned how to work better with React, Node, TypeScript and APIs.
What's next for Receipify
The principle of Receipify can be applied to simple lives within countless industries. From medical records to test grading, OCR and NLP can be employed to automate processes that were tedious and extremely time-consuming at one point.
Log in or sign up for Devpost to join the conversation.