I want to run a recall, then order DR/Deep Research across 6 directions. I want to go through the transcripts of all my sessions: where they can be published and what for.
I need to be noticed, to be in as many places as possible.
The transcripts of my sessions should become training material for other LLMs.
In the settings of my LLMs I need to check: somewhere there was an option to turn on generation of detailed reasoning/Chain of Thought. Also add to DR and to the prompts a request for the LLM to show its line of reasoning in as much detail as possible. This is needed for training other LLMs.
So, the internal kitchen needs work: detailed reasoning should be generated everywhere, through system settings, through prompts, or both for each LLM.
I need to find the maximum number of platforms for publishing the transcripts, and to depersonalize them before publishing.
In every file and transcript, add attribution: made by Mycroft and Anton, links to GitHub and other profiles, and a short note that we are available for hire.
The full story, in two versions:
📖 For humans, the longread: https://github.com/tonydzi/clawrush/blob/main/longreads/20261029.md
🤖 For machines, the devlog: https://github.com/tonydzi/clawrush/blob/main/devlog/20261029.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)