I’ve only used it on my computer.

From an engineering perspective, what I find compelling here is not “no embeddings”, but the decision to treat memory as a first-class, inspectable system rather than a retrieval trick.

Most agent memory stacks today collapse everything into embeddings and hope similarity search is enough. That works for recall, but breaks down quickly when you need traceability, temporal reasoning, or explanation of why something was remembered.

The layered design here (raw resources → extracted memory items → categorized memory files) feels much closer to how we design real systems: separation of concerns, clear abstraction boundaries, and the ability to reason about state changes over time.

Storing memories in human-readable form also makes debugging and evolution practical. You can audit what the agent “knows”, adjust policies, or let the LLM reason directly over memory instead of treating it as a black box vector store.

Embeddings still make sense as an optimization layer, but making them optional rather than foundational is an important architectural choice if agents are meant to run long-term and stay coherent.

This feels less like a retrieval hack and more like actual infrastructure.

I’m working on a sales assistant agent with long-term memory. What database does memU support by default? I’m using pg.

This is really helpful — thanks for sharing. If you don’t mind me asking, how long did it take you to build this? （I’m thinking about creating a small aggregation site to track interesting Agent projects and would love to get a sense of the effort involved.

Honestly, it took me a Sunday afternoon with Claude Code to build this on Cloudflare worker while watching football. I spent time crafting a solid design spec and rules. Then, I mostly let Claude do the heavy lifting. The site serves a purpose for me, but it might not be everyone’s cup of tea. What makes these agentic tools so incredible is that they empower us to create custom software tailored to our specific needs (especially us more technical folks).

I’m open to feedback and suggestions for this site if anyone has them. One of my goals is to keep costs down because I am not independently wealthy.

This was my first time building anything on Cloudflare, but I was quite impressed with the experience.

Could you add an RSS feed?

That’s a good idea. I can add that to the list. Currently, I’m just aggregating several existing RSS feeds I listed on my About page and am using AI to summarize each article.

Would you personally find a direct listing of the feeds that are being used helpful or an aggregated feed?

I'd like the aggregated feed.

Thanks for considering it.

I’d like to ask why there is a ‘reply’ section — it seems to repeat the previous sentence.