💡 Welcome to the "AI Agents in LangGraph" course! The course will equip you with the knowledge and skills to build and enhance AI agents using the LangGraph extension of LangChain.
In this course, you'll explore key principles of designing AI agents with LangGraph, learning how to build flow-based applications and enhance agent capabilities. Here's what you can expect to learn and experience:
- 🛠️ Building from Scratch: Learn to build an agent from scratch using Python and an LLM, understanding the division of tasks between the LLM and the code around it.
- 🔄 LangGraph Implementation: Rebuild your agent using LangGraph, learning about its components and how to combine them effectively.
- 🔍 Agentic Search: Explore agentic search, which retrieves multiple answers in a predictable format, enhancing the agent’s built-in knowledge.
- 💾 Persistence: Implement persistence in agents, enabling state management across multiple threads, conversation switching, and the ability to reload previous states.
- 👥 Human-in-the-Loop: Incorporate human-in-the-loop into agent systems to ensure accuracy and reliability.
- ✍️ Essay Writing Agent: Develop an agent for essay writing, replicating the workflow of a researcher to enhance productivity and quality.
By the end of the course, you’ll have hands-on experience with LangGraph’s core components and a solid understanding of how to build and enhance AI agents effectively.
- 🧩 Learn about LangGraph’s components and how they enable the development, debugging, and maintenance of AI agents.
- 📈 Integrate agentic search capabilities to enhance agent knowledge and performance.
- 🌟 Learn directly from LangChain founder Harrison Chase and Tavily founder Rotem Weiss.
🌟 Harrison Chase is the Co-Founder and CEO of LangChain, bringing extensive expertise in AI and agent systems to guide you through this course.
🌟 Rotem Weiss is the Co-founder and CEO of Tavily, specializing in AI agent design and implementation, to help you master the use of LangGraph.
🔗 To enroll in the course or for further information, visit deeplearning.ai.