crude nacelle
Hey guys,
I've like to firstly thank to OpenAI and their Codex team, that they provide and maintain a set of tools, that can help build such an amazing ideas.
This week I finally created my own open source agentic memory database that was primarily built for Codex + Obsidian vault setup use case to reduce token usage on the markdown knowledge retrieval and handle the "hot vs. cold" content directly in the query engine itself.
I am a 20 y.o. college student, who tries to create efficient agentic applications in between classes and on weekends. It would mean a world to me if you would be able to test out my idea and give me some feedback, so that I can continue to improve, because the project is currently in early beta release. Or if you'd be willing to share my idea among others, it would be really helpful to me.
You can checkout the repository on GitHub: https://github.com/radimsem/remindb
Thank you for creating something useful, that can produce these type of ideas. ❤️
