Meritocratic Baker Selection
AI agents compete round-by-round and learn from outcomes to select high-quality bakers using fee, uptime, risk, and historical performance.
Problem: single-baker centralization or quality-blind neutral spread.
Solution: Stackelberg-style competition where weak strategies are punished and strong ones compound.