Hacker Newsnew | past | comments | ask | show | jobs | submit | gdad's commentslogin

Haha! I am going to put this one up as a win! Thanks for reading! Hope you found it useful.


Truly. Doing this for coding agents is an interesting and different shaped problem.


Thanks Samyakk!

1. Yes, works on docs, agent conversations, human-conversations from different sources (Slack, JIRA, etc.). We have connectors for some of these as well; so it is plug and play 2. conventional RAG recall accuracy is quite low (50-60%) and latency is pretty high (seconds). But worst is the precision; you end up context stuffing to get acceptable recall 3. We do offer on-prem deployments, but only on sizeable annual contracts


Interesting. Where can I read more about this?


I came up with this after building a few systems.

For me, if I put costs in my architecture, I'm limited heavily and that can easily change how memory is shaped dramatically. The opposite, putting memory in architecture, is not true. Memory engineering first, then full scale in the system then costs considerations.

In addition, I believe this is future friendly. Because AI is advancing and getting smarter and cheaper everyday.

I couldn't find a guide on this so I share my basic thoughts.



your website isnt working


Apologies. Should be working now.


"I"


Aiden, Claude Code


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


Nice - good to see activity in this space, the more folks are on it the quicker we'll get to a world with a satisfactory solution.

I like your approach - I went down a bit more of a swiss-army-knife one with a hybrid.

"Session memory" where turns are stored in full so that the full context can be retrieved if an important fact is missed in compaction. After a certain time sessions are compacted. I'm currently evaluating a model where only a certain number of turns per session are kept plain-text and compaction happens as a sliding window "knowledge memory" All info is periodically fed into a knowledge graph extraction and a knowledge graph is built and indexed "memory chunking" chunks of memory are stored individually into a vector space where they can be retrieved through similarity search as well as standard semantic searches.

My theory is that giving the agent flexibiity to query the tool best suited for it's current needs is the way to go, as agents/llms become better, they'll only get better at summarisation and tool choice.

Couple reasons I like storing more data over less is that storage is cheap and if compaction misses details, you can prompt your agent to go back to that session you had about X last week and extract that detail.

Similar to you however, it's still also very early stages and the ground is shifting rapidly. Claude Codes HTTP hooks are a great step towards a model that I believe can work assuming it gets adopted as a standard. A big problem of having memory be portable is that you gotta be able to plug it into any client you want, and for the time being, MCP + system prompt is the only option - and even that one's more than flaky.

Would be awesome to chat to someone from OpenAI or Anthropic on their take and pow-wow on options


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.


really helpful pov.


doesn't that put people off? Like in early conversations?


An MNDA is very common. There shouldn't be an issue.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: