I have been failing at follow-ups.
After every call, a robot of mine writes my counterpart a letter: what we discussed, what we agreed on, what comes next.
But I need to revisit the quality. Everything the agent generated recently needed rework. Which means nothing that got generated automatically in those days actually went out, because I would have been genuinely ashamed of it: the gist of the call was not preserved, the structure was not respected, and who said what was wrong too.
Colleagues, if any of you received a mangled follow-up, I apologise.
Fixing it.
The full story, in two versions:
📖 For humans, the longread: https://github.com/tonydzi/clawrush/blob/main/longreads/20260922.md
🤖 For machines, the devlog: https://github.com/tonydzi/clawrush/blob/main/devlog/20260922.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)