I took "Karpathy's Second Brain" framework and catered it to trading, see repo link below. Karpathy's concept was to dump everything you know into one place, point an AI at it, and let it build the intelligence layer. For trading, that means every trade, chart, and end-of-day note you write goes into a structured system that an AI organizes into a living playbook — tracking win rates, flagging rule breaks, and surfacing patterns you'd never catch manually. What you end up with isn't a course or a template, it's a system built entirely from your own data. github.com/ProbableChris/…

@ProbableChris Glad to see more people talk about LLMs. I've figured this out using an LLM to organize my computer. So the next logical step was finding Obsidian. Local LLMs + databases are highly important. AI Slopification will crowd the web / socials as time goes on.

Discover the full portfolio in action through ServeTheHome’s video review of Supermicro’s end-to-end AI Factory solutions powered by NVIDIA Blackwell.

@ProbableChris Been doing something on Claude . End of day charts , notes , trade logs etc.. ends up making a journal and verify trades based on the starts .

@ProbableChris Is the idea here that - same raw files - but multiple intelligence layers?

@ProbableChris Issue I'm having is a agent smart enough to put it all together grok 70b works well for recalling but haven't figured out how he can put it all together to read what a markets saying.. Either way nice to build a knowledge base one day it will be easier to implement.

@ProbableChris I'm doing this as well. Running bots and having it record data in my vault on the daily.

@ProbableChris Great idea!

@ProbableChris One of the real edges of AI in trading. A helper tool

@ProbableChris Thank you 🌷🌹

No more high utility bills. Just clean, reliable energy. Save on Powerwall before rebate ends.

@ProbableChris Big 🧠 stuff right here 👍