DeepLearning.AI, a global edtech company, has released a free course titled “LLMs as Operating Systems: Agent Memory” to help people build AI agents. Students can learn how to build agentic memory into applications in this short course, LLMs as Operating Systems: Agent Memory, created in partnership with Letta and taught by its founders, Charles Packer and Sarah Wooders.

In this course, the students will learn how to build an agent with self-editing memory from scratch using tool-calling and multi-step reasoning. They will study Letta, an open-source framework that adds memory to LLM agents, giving them advanced reasoning capabilities and transparent long-term memory.

Based on the innovative approach in the MemGPT research paper “Towards LLMs as Operating Systems,” its authors, Charles and Sarah, proposed using an LLM agent to manage this context window, building a management system that provides applications with managed, persistent memory. 

This AI course will teach the students the key ideas behind the MemGPT paper, the two tiers of memory in and outside the context window, and how agent states comprised of memory, tools, and messages are turned into prompts. They will learn how to create and interact with a MemGPT agent using the Letta framework, how to build and edit its core and archival memory, how core memory is designed and implemented, and how to implement multi-agent collaboration.

By the end of the course, the students will have the tools to build LLM applications that can leverage virtual context, extending memory beyond LLMs' finite context window.