How Cisco Powers AI Automation with LangGraph, LangSmith, and LangChain | LangChain Interrupt

Video Description

Watch our recorded sessions from Interrupt here: https://interrupt.langchain.com/video/?utm_medium=social&utm_source=youtube&utm_campaign=q2-2025_interrupt-2025_co Discover the exact blueprint Cisco ($56B revenue) used to transform customer experience with AI agents—achieving over 95% accuracy. In this talk, Carlos from Cisco shares how 60% of almost 1.8 million support cases at Cisco are touched by AI-driven automation, many of those through LangGraph, LangSmith, and LangChain. What You'll Learn: - The 3-criteria framework Cisco uses to validate AI use cases - Why LLMs are "really bad for predictions" and how to combine them with ML models for enterprise accuracy - The "supervisor approach" that decomposes complex customer queries across specialized agents - Real production deployment strategies for on-premise, cloud, and hybrid environments - How to avoid the biggest mistakes in enterprise AI implementation Ready to build production agents? Get started with LangGraph for orchestration, LangGraph Platform for deployment, and LangSmith for observability. LangGraph Platform is now Generally Available: Deploy & manage long-running, stateful Agents: https://blog.langchain.dev/langgraph-platform-ga/