<p>This episode covers <strong>AIE World&#39;s Fair Recap of Day 2</strong> focusing on <strong>Keynotes &amp; SWE Agents</strong>. </p><p>🧠 Key Takeaways:</p><ul><li><strong>Moore’s Law for AI Agents:</strong> Capability is <strong>doubling every 70 days</strong>—yes, you read that right.</li><li><strong>Specifications = “New Code”:</strong> Aligning human intentions/values directly with model behavior—beyond old-school code artifacts.</li><li><strong>Evals:</strong> Absolutely critical for shipping AI, enabling rapid experimentation and tight feedback loops.</li><li><strong>Dagger “Container Use”:</strong> Secure, customizable, and multiplayer-ready agent environments.</li><li><strong>Thinking in Gemini:</strong> Models now iteratively “think” for smarter, dynamic responses with variable compute.</li><li><strong>Google Jules:</strong> Async coding agent supporting multitasking and parallel experimentation.</li><li><strong>GitHub Copilot Agent Mode:</strong> Autonomous searching, task execution, and self-healing for dev workflows.</li><li><strong>Brain Trust Loop Agent:</strong> Automated prompt, dataset, and scorer optimization—total eval game-changer.</li></ul>