The production App for the PDF Fact-Check Analysis platform. Built with Django, Python 3.13, and PostgreSQL, optimized with a stateless architectural design.
- Stateless PDF Processing: Zero physical storage writes for uploaded documents to guarantee enterprise-level data privacy.
- Concurrent Fact-Checking: Leverages multi-threading with optimized minimal thread pools to process extracted claims in parallel.
- Automated Live Search: Integrates with concurrent search endpoints to cross-reference statements across web indices dynamically.
The backend pipeline processes files fluidly using a strict RAM-only lifecycle:
graph TD
A[User Upload] --> B[1. Request In RAM]
B --> C[2. Database Entry]
C --> D[3. RAM Parsing]
D --> E[4. Garbage Cleanup]
style A fill:#2d3748,stroke:#4a5568,stroke-width:2px,color:#fff
style B fill:#1a202c,stroke:#4a5568,stroke-width:1px,color:#fff
style C fill:#1a202c,stroke:#4a5568,stroke-width:1px,color:#fff
style D fill:#1a202c,stroke:#4a5568,stroke-width:1px,color:#fff
style E fill:#1a202c,stroke:#4a5568,stroke-width:1px,color:#fff