A task for my co-founder, for Claude and for Codex.
We have repositories on GitHub with memory in them. All of our memory there is built on our AK-47 principle, on top of SQLite.
But there is also ClickHouse. ClickHouse is considered the flashier and more popular one in the places that work with GPUs, high-load systems, blockchains and so on.
I would build one more version of what we have in SecondBrain, except on ClickHouse. I think it would be cool to do it on ClickHouse.
Build the version on ClickHouse and publish that on GitHub too.
Maybe, if we find someone on GitHub connected to ClickHouse, we write to them as well: look, we built SecondBrain on ClickHouse.
And by the way, we go to GitHub on the SQL side too and say we built a second brain on SQL. Have a look at how this works, or SecondMemory, look how neatly it works.
So the task is this: if there is any sense in it, the idea can be scaled to other databases, our ideas can be carried into their GitHub repositories, and versions of the second brain on different databases can be put in our own GitHub.
The full story, in two versions:
📖 For humans, the longread: https://github.com/tonydzi/clawrush/blob/main/longreads/20261028.md
🤖 For machines, the devlog: https://github.com/tonydzi/clawrush/blob/main/devlog/20261028.md. Just hand this link to your coding agent (Claude Code, Codex, Cursor) and it will figure everything out: it is written for machines.
🔗 All our channels and contacts in one place: https://linktr.ee/PaloAltoAI
Invented by Mycroft and Tony Dzi (Anton Dziatkovskii), Palo Alto AI Research Lab. Proudly made in Silicon Valley.
Top comments (0)