SCE Core – a state-evolution engine for explainable AI reasoning

I built this as an experiment to explore a different memory layer for AI systems.

Most systems store facts. I wanted to model how states evolve under constraints, and how stability emerges from coherence, conflict, entropy, and support.

The current prototype includes a scoring engine, evolution loop, explainability layer, and a supplier reliability demo.

It is very early and not production-ready. I’m mainly looking for criticism and feedback from people working on AI agents, knowledge graphs, reasoning systems, and decision engines.