video1.0<iframe src="https://www.loom.com/embed/792ce1a6b73c4fc0bd0c5696a8bd6b60" frameborder="0" width="1316" height="987" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>9871316Loomhttps://www.loom.com9871316https://cdn.loom.com/sessions/thumbnails/792ce1a6b73c4fc0bd0c5696a8bd6b60-f214bac154b7701d.gif746.628In this video, I present my memory comparison project, where I explore three different memory strategies for a Langchain conversational agent: baseline, session memory, and long-term memory. The baseline strategy stores no information, while session memory retains context within a single thread, and long-term memory allows for cross-thread recall of key facts. I demonstrate these strategies through three scripted conversations, highlighting their distinct behaviors and storage mechanisms. I encourage you to consider how these memory strategies could be applied in real-world scenarios, such as customer support or personal AI assistants. Please take a look at the code and tests I implemented to see how these strategies function in practice.