Hello! First time posting here but wanted to share a project I’ve been working on in hopes others might find it useful as well. It’s a drop-in memory storage solution for LLM enabled projects that scales and solves the context size problem. In other words, it makes giving your AI features (such as chat) memory across sessions extremely easy! It uses vector stores and automatically handles storage and retrieval of information from past conversations. It also supports multiple LLMs (including GPT of course) and vector stores out of the box and I’ve added a lot of flexibility so if you prefer a different LLM or vector store you can easily implement your own. I’d love to hear what folks think, if there are any obvious issues of crucial missing features, and how you might use this in your own work!