PulseAugur
LIVE 11:27:02
tool · [1 source] ·
2
tool

Developer Rebuilds Operational AI System AxiomOS/Justicar

A developer is rebuilding AxiomOS/Justicar, an AI system that was operational and featured self-healing architecture, autonomous LoRA training, and constitutional governance. The system utilized a knowledge graph with 496 nodes and over 5,000 edges, employing Mixture-of-Experts routing across models of 3B, 7B, and 14B parameters. It was built on a small $600 PC and was lost due to hardware failure and a medical crisis, but the architecture is being redeveloped with documentation and GitHub repositories now available. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Details the rebuilding of a complex AI system, offering insights into its architecture and development process.

RANK_REASON The cluster describes the rebuilding of a specific AI system, which falls under the category of a tool or project rather than a major industry release or research milestone.

Read on Mastodon — fosstodon.org →

Developer Rebuilds Operational AI System AxiomOS/Justicar

COVERAGE [1]

  1. Mastodon — fosstodon.org TIER_1 · [email protected] ·

    Rebuilding — @Phenarax-ui AxiomOS / Justicar wasn't a prototype. It was running. Self-healing architecture. Autonomous LoRA training. Constitutional governance.

    Rebuilding — @Phenarax-ui AxiomOS / Justicar wasn't a prototype. It was running. Self-healing architecture. Autonomous LoRA training. Constitutional governance. Knowledge graph: 496 nodes, 5,000+ edges. MoE routing across 3B/7B/14B models. Meta-teacher generating its own training…