"A customs officer at the Tunisian tax authority was spending entire days manually searching through thousands of scanned PDFs to find a single tariff ruling. I built DouaneXtract to solve this β a full-stack, self-hosted platform that indexes 3,007+ documents and returns results in under 1 second."
# π DouaneXtract
> **Tunisian Customs Document Intelligence Platform**
> Transforming thousands of scanned PDFs into an instantly searchable, secure knowledge base.
---
## π― The Problem
A customs officer at the **Tunisian Customs Authority (Douane Tunisienne)** was spending entire days β sometimes multiple days β manually searching through thousands of scanned PDF documents to find a single tariff ruling or administrative notice.
**No search. No structure. No time.**
---
## π‘ The Solution
I designed and built **DouaneXtract** from scratch as a **solo freelance project** β handling everything from requirements gathering to production deployment.
A self-hosted, secure web platform that:
- Indexes **3,007+ scanned documents** across 3 categories
- Returns full-text search results in **under 1 second**
- Provides **role-based access control** so sensitive documents stay protected
- Runs entirely on a **private server** β no cloud dependency, no data leaks
---
## πΈ Screenshots
| Login | Dashboard | Search |
|-------|-----------|--------|
| | | |
---
## π Impact
| Metric | Value |
|--------|-------|
| π Documents indexed | **3,007** |
| β‘ Search response time | ** *"From a frustrated customs officer spending days searching PDFs, to finding any document in under a second."*
---
## π¬ Contact
**Chourouk** β Freelance Developer & DevOps Engineer
---
*Built with β€οΈ for the Tunisian Customs Authority*