Optimizing AI Agents with ART: Reinforcement Learning for Smarter Decisions

Artificial intelligence agents are rapidly evolving, but how do we make them learn better behaviors in complex real-world environments? In this session, we’ll explore **ART (Agent Reinforcement Training)** from **OpenPipe**, a framework that uses reinforcement learning (RL) to refine how AI agents make decisions, handle uncertainty, and improve over time. We’ll break down: * What ART is and why it matters for agent performance * How reinforcement learning techniques can improve reasoning and adaptability * Practical examples of training agents with ART using OpenPipe * Real-world use cases: from customer support agents to autonomous workflows This event is ideal for engineers, researchers, and startup builders who want to push beyond static prompt engineering and learn how to apply reinforcement signals to build smarter, more reliable AI agents. Join us for an interactive demo and discussion on how ART can transform your AI development workflow.