---
title: PaperPilot
emoji: 📄
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: "6.16.0"
python_version: "3.10"
app_file: app.py
pinned: false
tags:
  - track:backyard
  - sponsor:openbmb
  - sponsor:openai
  - sponsor:nvidia
  - sponsor:modal
  - achievement:offbrand
  - achievement:fieldnotes
---

# 📄 PaperPilot

AI-powered Scholarship & Form Assistant

**PaperPilot** helps students and applicants understand lengthy application forms instantly by extracting key information, checking eligibility, generating document checklists, and answering questions using AI.

---

# 🚀 Problem Statement

Many students and applicants struggle with:

* Long and complex application forms
* Hidden eligibility requirements
* Missing deadlines
* Missing mandatory documents
* Confusing instructions

PaperPilot solves this by automatically extracting and explaining the most important information from uploaded documents.

---

## Solution

PaperPilot uses OCR and AI to analyze forms and provide structured, easy-to-understand information.

Users simply upload a PDF and PaperPilot automatically extracts:

* Form Name
* Deadlines
* Eligibility Rules
* Required Documents
* Contact Information
* Form Summary

It also provides an AI assistant for asking questions about the uploaded form.

---

# ✨ Features

### 📄 Smart PDF Processing

* Supports normal PDFs
* Supports scanned PDFs
* OCR-based text extraction

### 🤖 AI-Powered Form Understanding

* Automatic form analysis
* Structured information extraction
* Form summarization

### ✅ Eligibility Analyzer

* Detects eligibility criteria
* Extracts income limits
* Identifies category requirements
* Highlights important conditions

### 📋 Document Checklist Generator

* Extracts required documents
* Generates actionable checklists

### 📅 Timeline Extraction

* Detects deadlines
* Highlights important dates

### ⚠ Risk Detection

* Identifies missing information
* Detects critical deadlines
* Warns users about possible issues

### 🔍 Document Verification Assistant

* Helps users verify required documents before submission

### 💬 Ask PaperPilot

Powered by **Qwen 2.5 Instruct LLM**

Users can ask natural-language questions such as:

* What is the last date to apply?
* Am I eligible?
* What documents are required?
* Summarize this form.
* What happens if I miss the deadline?

---

# 🏗 System Architecture

```text
User Uploads PDF
        │
        ▼
OCR Engine
(Normal + Scanned PDFs)
        │
        ▼
Text Extraction
        │
        ▼
Master JSON Builder
        │
 ┌──────┼──────┐
 ▼      ▼      ▼
Summary Eligibility Checklist
 │        │        │
 ▼        ▼        ▼
Timeline Risk Detection Verification
        │
        ▼
Qwen 2.5 AI Assistant
        │
        ▼
User-Friendly Insights
```

---

# 🛠 Tech Stack

### Frontend

* Gradio

### Backend

* Python

### OCR

* EasyOCR
* PyMuPDF
* PDFPlumber

### AI / NLP

* Hugging Face Inference API
* Qwen 2.5 Instruct

### Deployment

* Hugging Face Spaces

### DevOps

* GitHub Actions
* CI/CD Pipeline

---

# 🔄 CI/CD Pipeline

```text
GitHub Repository
        │
        ▼
GitHub Actions
        │
        ▼
Hugging Face Spaces
        │
        ▼
Automatic Deployment
```

Every push to the main branch automatically deploys the latest version of PaperPilot.

---

# 🎯 Use Cases

* Scholarship Applications
* Government Schemes
* Admission Forms
* Internship Applications
* Job Applications
* Grant Applications
* Registration Forms

---

## Demo Video

https://youtu.be/EalpBFBLPA0?si=UVeDujcsYHECKKSm

## Social Media Post

https://x.com/KamranX07/status/2066492830475038881

## GitHub Repository

https://github.com/KamranX07/PaperPilot

---

## 📝 Field Notes

### Problem
Students often struggle to understand scholarship and application forms due to complex eligibility rules, deadlines, and document requirements.

### Solution
PaperPilot is an AI-powered assistant that analyzes forms, extracts important information, verifies eligibility, and answers user questions in natural language.

### Technical Architecture
- OCR/Text Extraction
- Structured Form Parser
- Master JSON Generation
- Eligibility Verification Engine
- Qwen 2.5 7B Instruct via Hugging Face Inference
- Gradio Frontend

### Challenges Faced
- Hugging Face legacy inference endpoints were deprecated.
- Migrated from the old API route to the new InferenceClient-based routing system.
- Resolved dependency conflicts between Gradio and Pydantic.
- Improved eligibility extraction logic for income limits.

### Lessons Learned
- Small language models can solve practical real-world problems effectively.
- Structured JSON extraction greatly improves reliability.
- Good UI/UX significantly improves user adoption.

### Future Enhancements

* Multi-language support
* Advanced eligibility reasoning
* Personalized recommendations
* Form autofill assistance
* Voice-based interaction
* Mini-RAG document memory
* Mobile application

---

# 👨‍💻 Team

- Hugging Face Username: KamranX07 (Solo)

Built as an AI-powered document intelligence platform for simplifying form understanding and application workflows.

---

# ⭐ PaperPilot

**Upload. Understand. Apply with Confidence.**
**Built for the Build Small Hackathon.**