Building Agentic and Multi-Agent Systems with LangGraph

This session breaks down the concepts and code needed to understand and build industry-standard agentic and multi-agent systems using LangChain and LangGraph. The talk covers prompt engineering, Retrieval Augmented Generation (RAG), and agentic patterns, emphasizing how LangGraph enables cyclical reasoning loops and stateful applications. It demonstrates building an agentic weather application using tools like DuckDuckGo search and a custom weather tool, showcasing the integration of these components within a graph structure. The presenters highlight the importance of defining tools with clear descriptions for LLMs and the role of LangChain Expression Language (LCEL) in composing these applications for production.