Introducing Aperture Nexus: Open Source Memory for AI Agents, Built on ApertureDB

An agent walks into a repeat problem like it's the first time, every time. Most memory tools treat everything as disconnected text chunks in a vector store, so nothing carries over, not the reasoning, not the context, not what actually happened last time.

Aperture Nexus is our answer. Context, who, what, when, why, and how, is stamped on every commit, so retrieval actually means something later. Enable lineage tracking when you need it, and every commit traces back to its original source. And because it's built directly on ApertureDB (), the unified graph-vector-multimodal database already running in production, it works the same whether your agents are text-only today or need images, documents, video, or structured records tomorrow. No migration later if that's where you end up. Knowledge and Memory both live together in ApertureDB.

Most agents already have their own reasoning, Claude knows what's worth persisting as well as anyone does. This version of Nexus is built around that: a place for whatever reasoning is already happening to commit, retrieve, and build on, rather than introducing a competing reasoning layer of its own. Future versions will build on that foundation based on what we learn across different verticals.

This didn't come from a whiteboard. We audited over 20 existing memory frameworks to see what actually holds up, and shaped this against what we've learned deploying ApertureDB across organizations at very different stages of AI adoption. The thinking behind it: and . We also started a podcast on this exact question, The Cognitive Layer: .

v0.1, live now: commit, search, connect, authenticate.

What's next, public in the repo's discussions: Memory.update() with lineage, query-time visibility enforcement, GraphRAG neighborhood search, hybrid search, recency weighting, an MCP server.

It's open source, MIT licensed.

Try it:
Docs:

Try it, and if it solves something you're dealing with, star it or send it to someone hitting the same wall.

Do share if something breaks when you use it or what else you would wish Nexus would support.

Regards,

Vishakha

97 views

Add a comment

Replies

Best

"We audited over 20 existing memory frameworks to see what actually holds up, and shaped this against what we've learned deploying ApertureDB across organizations at very different stages of AI adoption."

I love that you started with such detailed analysis by auditing existing frameworks and saw what worked and did not work then built an open sourced tool. Sounds really interesting, and I saw this as someone who's build / abandoned (not quite abandoned but rather rolled into a different product) a memory tool in the past for AI.

 thank you. Would love to get feedback on what problems you can solve with it today and what you would like to see in future releases, especially since you have worked on this problem before!

I'm excited to try this out. It makes a lot of sense to let people incorporate any model/reasoning layer they want, and focusing more on enhancing the infrastructure around it and playing to the strengths of Aperture as a multimodal DB.

 thank you. Do share your thoughts once you get a chance to build with this.