SecondBrain - Academic Knowledge Assistant
Local-first, privacy-focused RAG system for students to build personal knowledge bases from lectures, PDFs, and videos. Features FAISS + BM25 hybrid search with academic metadata extraction.
Technologies Used
About This Project
Privacy-focused academic knowledge management system operating 100% locally for students. Implements hybrid search combining FAISS vector similarity with BM25 keyword search for optimal retrieval. Features academic metadata extraction (course, topic, professor), SQLite for metadata management, math-friendly embeddings (intfloat/e5-small-v2), and clean modular architecture. Planned features include Whisper-based video transcription, automated flashcard generation, and citation management. Built for students prioritizing data privacy while maintaining powerful RAG capabilities for studying and research.
Key Features
- • Feature 1: Add your project features here
- • Feature 2: Describe the main functionality
- • Feature 3: Highlight what makes it special
Challenges & Solutions
Describe the challenges you faced during development and how you solved them. This helps showcase your problem-solving skills.