Show HN: BrainAPI's event-centric graph from any data with four-stage pipeline

Category: ai-ml

Tags: knowledge-graph, agent-swarm, nlp, graph-database, event-centric

Score: 7.5/10 (Innovation: 7, Technical: 8, Documentation: 8, Utility: 7)

BrainAPI transforms unstructured text into an event-centric knowledge graph using a multi-agent pipeline (Scout, Observer, Architect, Janitor) that extracts entities, relationships, and context. Its innovative triangle-of-attribution model enables traceable, multi-hop question answering with explicit graph paths rather than similarity-based guesses. The project combines LLM agents, graph databases, and a user-friendly TUI/console to make complex knowledge graph construction accessible.

Target audience: backend devs, data engineers, ai researchers

Repository: https://github.com/Lumen-Labs/brainapi2 · Python · NOASSERTION · 132 stars

View on Hacker News