Show HN: Aurra – Bi-temporal memory for AI agents (with LLM auto-supersede)

Category: infrastructure

Tags: ai-agents, memory-infrastructure, bi-temporal

Score: 6.5/10 (Innovation: 6, Technical: 6, Documentation: 7, Utility: 7)

Aurra provides bi-temporal memory infrastructure for AI agents, enabling versioned fact storage with automatic supersession detection using LLM classifiers. The project is interesting because it tackles the practical challenge of stale agent memories with a novel combination of temporal versioning and semantic conflict resolution.

Target audience: backend devs, data engineers

Repository: https://www.aurra.us/blog/level-2-auto-supersede-beta · Python · MIT

View on Hacker News