Module 2
Short-term Memory
Short-term memory is the conversation itself — the sequence of messages between a user and an agent, stored as it happens and read back on the next turn.
In this module, memory moves from the shared workshop instance into your own, and your agent starts using it: you configure the client, give the agent its dependency architecture, and store and recall every turn. The optional lesson opens the API behind it, in the tour notebook.
By the end of this module, you will be able to:
- Point memory at your own instance, alongside the course Full definition for knowledge graph (opens in a new tab)A representation of real-world entities and their relationships, stored according to organizing principles, typically in a graph database.
- Configure the memory client with connection, Full definition for embedding (opens in a new tab)Information represented as a numerical vector, positioned so that similar information sits close together., and extraction settings
- Structure the agent with dependencies and a dynamic system prompt
- Store each turn, and recall the conversation across restarts