README.md

---
title: Cortex
emoji: 🧠
colorFrom: yellow
colorTo: red
sdk: docker
app_port: 7860
pinned: false
---

# Cortex — Multimodal Knowledge Intelligence

Upload PDFs, images, videos, and text files — ask questions across all of them.

## Features
- **Multimodal ingestion** — PDF, images, video frames, plain text
- **Semantic search** — sentence-transformers + FAISS vector index
- **Q&A with citations** — Claude API answers with source + page references
- **Auto summaries** — per-file summaries and entity extraction on upload
- **Cross-document connections** — Claude finds relationships between files

## Stack
- Backend: FastAPI + sentence-transformers + FAISS + PyMuPDF + OpenCV
- LLM: Claude API (claude-sonnet-4-6)
- Frontend: Custom HTML/CSS/JS (earthy professional design)
- Deployment: Docker Space on Hugging Face

Built by Yashika Saxena · B.Tech AI & ML · ITM Gwalior · 2023–2027