🐝 The First Self-Improving agents with RL / Prompting Optimization
👩⚖️ Agent-as-a-Judge: The Magic for Open-Endedness