Show HN: Open-Source Document Extraction Tool (github.com)

13 points by harishd30 ↗ HN
Hi, I'm building an open-source self-hostable document extraction tool powered by LLM.

There are popular data extraction tools and OCR tools in the market. None of them are open source.

Most accounting firms, law firms, insurance, back office, and real-estate folks would like to use a tool like this.

You can add PDF documents, Images, and audio files and create columns to answer questions on documents or extract information into tabular format.

Access to repo: https://github.com/harishdeivanayagam/rowfill Screenshots: https://github.com/harishdeivanayagam/rowfill/tree/master/sc...

Sample use-case: You can upload your invoices or receipts, create custom columns, and click Start Run. The data would be automatically extracted.

2 comments

[ 26.1 ms ] story [ 147 ms ] thread
FYI - website seems broken, docs link is invalid, readme mentions docker-compose but the compose file is missing, pnpm run build:server is broken, pnpm run start:app launches the app but the login fails as it seems it needs a database server, dockerfile used outdated node.
Thanks for the info. The project is relatively in the nascent stage. Will sort out the issues sooner.