Agentic AI-guided evaluation system for comparing LLMs with multi-judge jury scoring - View it on GitHub