PDF Rule Checker – Fullstack Assignment

This is a small fullstack project where the user uploads a PDF and enters 3 rules.
The backend checks the document against those rules using an LLM and returns:

• pass/fail
• short evidence sentence
• your reasoning
• confidence (0–100)

The goal was to keep things simple and readable.

---

Tech Stack

Frontend: React + Vite
Backend: Node.js + Express
PDF Parsing: pdf-parse
LLM: OpenAI-compatible API

---

How to Run

1. Backend

• cd backend
• npm install
• cp .env.example .env
• Add your OPENAI_API_KEY in the .env
• npm start
• Backend runs on http://localhost:4000

---

2. Frontend

• cd frontend
• npm install
• npm run dev
• Frontend runs on http://localhost:5173

---

Features

• Upload a PDF (2–10 pages)
• Enter up to 3 rules
• Clean table with pass/fail, evidence, reasoning & confidence
• Simple readable backend code

---

Screenshot

(Add your screenshot here once you run it locally)

---

Notes

The project uses a simple approach to PDF text extraction and confidence handling.
If needed, page-wise extraction can be added later, but for this assignment the current setup works well.