Developer Forks Archived Graph Database to Build Healthcare AI Analytics Engine

A software engineer has forked Kùzu, an archived embedded analytical graph database, and renamed it Gorgonzola, adding vector search, LLM embedding support, and graph algorithm extensions. The developer clarified that the core engine — including columnar storage, OpenCypher query processing, and ACID transaction management — was built entirely by the original Kùzu team. His own contributions include C++ API extensions, parser tweaks, integration of third-party HNSW vector indexing, and connectors to external LLM APIs such as OpenAI and Google Gemini. The project also wraps existing graph algorithms like PageRank and Louvain into an extension system, with AI coding assistants used heavily to manage the large C++ codebase. The developer describes Gorgonzola as a learning project with a long-term goal of making it a foundational engine for clinical analytics and healthcare AI applications.