LangGraph

• It helps managing the memory https://docs.langchain.com/oss/python/langgraph/memory
• Example: supports trimming old messages
• In LangGraph, the flow (composed by the steps in the application) is represented by a graph (or a state machine)
• Flow Engineering: having a graph to guide the AI application on the steps to take

Levels of autonomy in LLM applications

• https://blog.langchain.com/what-is-a-cognitive-architecture/

• LangGraph and the ReAct Agent are positioned as a State Machine

LangGraph components

• Nodes: functions that receive the GraphState and returns the updated keys on the state. Has the start and end nodes
• Edges: connect nodes
• Conditional Edges: connect nodes conditionally