Memory-Os: AI-Powered Personal Knowledge Operating System

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

Memory-OS is an AI-powered Personal Knowledge Operating System designed to unify personal and development-related information through a conversational command-line interface. The system integrates external knowledge sources such as GitHub, Gmail, and Notion through a connector-based architecture and processes collected information for efficient retrieval. It uses a local sentence-transformer model to generate embeddings, Qdrant for vector storage, Neo4j for knowledge graph relationships, and SQLite for structured data management. A hybrid Retrieval-Augmented Generation (RAG) pipeline combines semantic search, keyword-based retrieval, and graph-based context to generate relevant and context-aware responses using a Groq-powered Large Language Model (LLM). Memory-OS also provides an interactive terminal interface, a full-screen Textual TUI, workspace management, incremental synchronization, Docker-based service provisioning, health diagnostics, backup and restore functionality, and performance monitoring. The primary objective is to reduce information fragmentation by enabling users to search, connect, and understand their accumulated knowledge through natural-language queries instead of manually navigating multiple applications.