

Sleep Money Maker
4.4K posts

@SleepMoneyMaker
Ph.D. 20+ years building AI to see patterns humans miss. Pointed one at the options market. I post what it finds, how AI improves it, and weekly results.



- Drafted a blog post - Used an LLM to meticulously improve the argument over 4 hours. - Wow, feeling great, it’s so convincing! - Fun idea let’s ask it to argue the opposite. - LLM demolishes the entire argument and convinces me that the opposite is in fact true. - lol The LLMs may elicit an opinion when asked but are extremely competent in arguing almost any direction. This is actually super useful as a tool for forming your own opinions, just make sure to ask different directions and be careful with the sycophancy.

My AI agent stack right now: Paperclip (@dotta) - Orchestration Gstack (@garrytan) - Strategy Autoresearch (@karpathy) - Exploration All free, all open source.



We need a maestro to orchestrate a research community of autoresearcher agents


Okay let's see who can reply to this

Introducing TurboQuant: Our new compression algorithm that reduces LLM key-value cache memory by at least 6x and delivers up to 8x speedup, all with zero accuracy loss, redefining AI efficiency. Read the blog to learn how it achieves these results: goo.gle/4bsq2qI

I now steer my @karpathy autoresearch agent over Telegram. It reports every experiment as it lands: improved, reverted, or crashed. I read the results, think and /steer a new direction into program.md Agent picks it up next iteration. I go back to whatever I was doing. One Python file (under 630 lines of code). Called it labmeet.




You'll get addicted if you try this just once: /gstack run office hours on my codebase (h/t @garrytan)




