Firat Adem Bilge

Cognify is a full-stack personal knowledge management system — a second brain that actually reads your documents. Instead of manually tagging and filing notes, you drop in scanned PDFs, images, or plain-text files and an asynchronous Celery worker pipeline handles the rest: OpenCV preprocessing and Tesseract OCR extract text, DistilBART generates concise summaries, a zero-shot MNLI classifier auto-tags topics, and Sentence-BERT stores semantic embeddings — making every document instantly searchable and analyzable.

On the frontend, a Next.js 16 dashboard provides a knowledge vault overview, Recharts-powered analytics, browser-native PDF tools (split, merge, reorder), document versioning with parent-child lineage, per-document sensitive mode, soft-delete recovery, and a global RAG chat where RoBERTa generates cited answers from the most relevant chunks in your vault. The Django REST backend enforces JWT authentication, user-scoped data isolation on every query, audit logging for all actions, and Prometheus metrics for worker latency.

Key Capabilities

• Full ingestion pipeline: upload → OCR → summarize → classify → embed → COMPLETED
• Cosine similarity search finds meaning across documents, not just keyword matches
• Collections, annotations, bookmarks, and notes attached to any document or quote
• 35 automated tests, Docker Compose orchestration, MIT licensed

Tech Stack

Python Django Next.js Celery Transformers Sentence-BERT Docker Prometheus