Building in the same space (Maximem Vity, cross-LLM memory as a Chrome extension and OpenClaw plugin). The SMTP analogy really resonates with me. I wrote a comparison a few weeks ago of how ChatGPT, Claude, and OpenClaw actually implement memory under the hood [1] and the architectures are so different that interop feels almost accidental when it works.
One thing I keep running into: the hard part isn't storage or retrieval. It's qualification. Deciding what's worth remembering from a conversation vs. what's throwaway context. ChatGPT takes the "summarize the last 15 chats" approach, Claude does on-demand search, and both have real failure modes. We went with a semantic graph that tries to capture relationships between memories (your preference for serverless connects to your AWS project connects to your cost constraints) rather than flat key-value pairs. Still iterating on it honestly.
Re: Claude's import-memory launch yesterday, I think the timing validates the category but the approach is fundamentally migration, not sync. You import once, and from that moment your contexts diverge again. Anyone using 3+ tools daily (which is basically everyone I talk to) is back to fragmented memory within a week.
Curious about your retrieval approach. Are you doing hybrid search or pure semantic?
[1] https://www.maximem.ai/blog/ai-apps-memory
Author here. I have been learning about how ChatGPT, Claude, and OpenClaw handle persistent memory; what actually gets stored, how it is retrieved, etc.
Quick summary of the findings:
- ChatGPT pre-computes lightweight summaries of your ~15 most recent conversations and injects them into every prompt. No vector search, no RAG. Simpler than I expected.
- Claude takes an on-demand approach; it has search tools it can invoke to query your past conversations, but only fires them when it judges the context is relevant. More flexible, less consistent.
- OpenClaw stores memory as plain Markdown on your local machine with hybrid search (semantic + BM25, 70/30 weighting). Fully transparent, but single-platform.
Full disclosure: I'm building in this space (Maximem Vity — a private, secure vault for cross-LLM memory). The comparison stands on its own, but that context motivated the research.
Happy to discuss the architectural differences or answer questions.
I experienced this as well. The built-in memory features are a black-box. Can't see what's stored, can't control when it's recalled and takes random irrelevant examples from past conversations into unrelated conversations. Also, I am into using multiple models, so it was frustrating they can't talk to each other.
The .md file approach is cool but like you mentioned, it is an overhead and doesn't scale well.
I have been building something in this space (Maximem Vity [https://maximem.ai]) that tries to solve this more systematically: a cross-LLM, cross-app memory layer that sits in a secure cloud vault. The idea is you control what gets stored and you can summon specific pieces of it into any AI session, granularly. So instead of hoping GPT remembers that you prefer TypeScript and work at a fintech startup, you explicitly pull that context in wherever you need it.
I tried it last Monday and was in awe. All my team members now use it. I for one, am stepping cautiously around it. And hence had similar questions.
Yesterday, we released a memory plugin for it (based on their plugins framework). And we got 1000+ installs in 12 hours. My first instinct was that these must be bots. But our dashboard shows real-usage. So yeah, real users are there.