invoiceIQ

• 

  LOGO

Inspiration

When looking at business data, we found that multiple businesses loose around $10,000 a month alone due to invoice mishaps. So we took this liberty into our own hands by creating Invoice-Guard.

What it does

You first need to log in with company account, this will lead you to your dashboard. On your dash board you have many options to you such as seeking your invoice history, watchlist, analytics, and the most important the file upload. Once you upload your file or paste in your text our software take the info to our backend where our LLM (trained with gemini api) will detect key aspects of the invoice such as, the IBAN#, Tax#, whether the invoice has been paid or not, etc.

How we built it

Frontend: react + vite Backend: Node.js + Express.js Database: SQLight 3 OCR: Tesseract API LLM Model: Gemini API

Challenges we ran into

Training the LLM Model to recognise and save the data from past invoice for future invoices. Had a difficult time trying to add all the proper features and formats in the UI/UX design.

Accomplishments that we're proud of

Finding a an idea that has positive real world impact and scalability to become a software startup.

What we learned

Version Control, How to fully structure a full stack with connecters using GitHub.

What's next for Invoice-Guard

Getting funding with VC and investors to scale on a bigger level so we can grow and create our ML model.

Built With

• api
• computervision
• gemini
• javascript
• knn
• llm
• machine-learning
• node.js
• python
• react
• sqlight3
• sqlite
• tesseract
• vite