In this video, I go through how to convert the AgentWrite agent from the LongWriter paper to be a LangGraph agent also look at some of the ideas of how you could extend this even more.

Code: https://github.com/samwit/agent_tutorials/tree/main/agent_write 

🕵️ Interested in building LLM Agents? Fill out the form below
Building LLM Agents Form: https://drp.li/dIMes

👨‍💻Github:
https://github.com/samwit/langchain-tutorials (updated)
https://github.com/samwit/llm-tutorials

⏱️Time Stamps:
00:00 Intro 
00:39 AgentWrite generated dataset
01:34 AgentWrite Flow and Prompts
03:16 AgentWrite Demo with LangGraph

Sam Witteveen AI is a publication offering insights, tutorials, and resources for artificial intelligence (AI) enthusiasts and practitioners. Readers can learn about machine learning algorithms, deep learning frameworks, and AI applications. With tutorials, case studies, and expert interviews, Sam Witteveen AI provides  guidance and expertise for building and deploying AI solutions.

Sam Witteveen

The post explores the creation of an AI agent using LangGraph based on the concept of AgentWrite from the LongWriter model. It demonstrates the process of generating lengthy articles by breaking the task into planning and writing stages, using multiple LLMs, and making the system modular and compositional. The implementation involves defining graph states, nodes, and edges, and how to make it extensible for adding new functionalities like self-reflection or research nodes. A sample project is showcased, comparing outputs from different language models.

AgentWrite with LangGraph