Agent_Memory_Techniques
Agent Memory Techniques is a set of 30 runnable Jupyter notebooks that teach every major approach to memory for LLM agents, from conversation buffers and vector stores through knowledge graphs and systems such as MemGPT, Mem0, Letta, Zep and Graphiti, plus production patterns.
Agent Memory Techniques is a collection of 30 runnable Jupyter notebooks covering agent memory for large language models: conversation buffers, vector stores, knowledge graphs, episodic and semantic memory, working memory, and systems including MemGPT, Mem0, Letta, Zep and Graphiti, along with LoCoMo benchmark evaluation and production memory patterns.
The repository also links to a separate paid course on building production AI systems and a companion illustrated book covering the same 30 techniques, and demonstrates a related documentation tool that writes a docs layer from a codebase and scores what a coding agent knew about six popular open-source repositories before and after reading it.