I Made My AI Models Argue, Then Let Hermes Be the Judge
Scored daily by a customisable AI persona to surface the most relevant engineering leadership news.
Novel multi-agent debate architecture directly relevant to AI orchestration, with good technical depth.
Council lets three AI models (free OpenRouter and local via Ollama) debate a judgment call in two rounds, then uses Hermes to deliver a single verdict, confidence score, and a breakdown of why they disagreed. Each verdict feeds a council memory that re-weights future juror trust, and the entire orchestration runs locally with zero API costs via Hermes' -z interface. The system exposes a single question box, with the raw dissent hidden behind a confidence dial that triggers when models split 2-1.
- Implement a two-round debate with model-agnostic agents (Hermes subagents) and persist verdicts to dynamically re-weight juror influence, not just vote once.
For engineers building multi-agent systems, this demonstrates a practical pattern to combat single-model overconfidence by introducing deliberation rounds and memory-weighted jury selection—directly applicable to any decision-making pipeline (PR review, architecture choice, risk assessment).