A local AI that types up a Brazilian accountant's receipts (and knows when it's wrong)
This is a submission for the Hacktoberfest Open-Source AI Challenge: Week 2 — Side Hustle . What I Built A self-employed accountant in Brazil gets ~100 documents a week from her small-business clients: NF-e invoices (PDF), NFC-e receipts photographed on a phone, Pix transfer screenshots, and the occasional handwritten recibo . Today she types every one of them into a spreadsheet by hand. Leitor…
A Brazilian accountant utilizes an open-source AI to process her clients' invoices, receipts, and Pix transfers. This software, called "Leitor de notas," converts the incoming documents into an Excel file for review. The AI runs locally using the Ollama framework, enabling privacy and eliminating the need to upload sensitive financial data to the cloud.
The system reads each document with an open-weight model and validates the results before exporting them to Excel. This approach saves accountants valuable time and eliminates the need for costly API fees. The AI can handle various document types, such as PDFs, images, and handwritten receipts, and it remains accurate even when faced with challenging scans.
Ollama's local processing ensures the information stays within the accountant's system, adhering to Brazil's data protection law (LGPD).
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
