Zero-Cost, 100% Private: Building a Local Invoice & Receipt Sorter with Open-Source AI

This submission aims to win the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle, presenting an offline, zero-cost Python invoice and receipt sorter called "Local Invoice & Receipt Sorter." The tool is designed to automate expense tracking for freelancers and side hustles by processing financial documents locally, categorizing vendors, and producing structured JSON files—without transmitting any data to cloud-based APIs or third-party servers.

The system operates entirely offline, scanning your local directory and parsing receipts instantly. It leverages a lightweight Python script that utilizes open-source libraries and local file system traversal. The core of the application resides in the `parse_local_invoice` function, which parses invoices locally using open-weight vision/text models, ensuring all financial data remains secure on your personal device.

Here is how the pipeline functions:

1. The script starts by listing all files in the designated invoices directory (e.g., `./unprocessed_invoices`).

2. It identifies files with common document extensions such as `.pdf`, `.png`, and `.jpg`.

3. For each identified file, the `parse_local_invoice` function is called, which processes the file and returns a dictionary containing essential details like the vendor name, date, total amount, and category.

4. The output is then printed in JSON format, making it easy to read and utilize.

The software harnesses the power of open-source technology, ensuring zero recurring costs and absolute data privacy. By processing all financial data locally, the tool circumvents the need for token-based pricing models or monthly subscription fees, which are common in proprietary SaaS applications. This approach also eliminates the risk of vendor lock-in, granting users complete ownership over their automation pipelines and financial data.

An early validation of the tool was conducted with a freelance design client, who processed monthly software subscriptions and equipment receipts entirely offline. Feedback confirmed the system successfully maintained zero data leakage concerns and efficiently organized unstructured invoices into structured JSON files without any recurring API subscription fees.