God of Prompt
RT @godofprompt: This was 6 years ago.

Wondering how we’ll look back on clawdbot 6 years from now on.

I just built a *functioning* React app by describing what I wanted to GPT-3.

I'm still in awe. https://t.co/UUKSYz2NJO - Sharif Shameem
tweet

God of Prompt
RT @godofprompt: codex is used by all top engineers

codex hallucinates less and is more reliable than claude code

at least for now

Introducing the Codex app—a powerful command center for building with agents.

Now available on macOS.

https://t.co/HW05s2C9Nr - OpenAI
tweet

God of Prompt
RT @godofprompt: you’re all lying about openclaw

you’re leaving it run whole night, going safely to sleep

because deep inside you know it’s not AGI nor it’s sentient

when it will be AGI for real, you’ll never sleep the same way again
tweet

Moon Dev
clawdbot is cool, but can it make you money?

this guy had his clawbot find him a 71,057% ROI trading strategy

the complete masterclass: https://t.co/BmQ34aMcmF
tweet

Javier Blas
OIL MARKET: Shell CEO Wael Sawan highlighting how little oil prices have moved (around $60-$70 a barrel) despite the huge increase of geopolitical risk and the use of sanctions. The market is far more stable than in the past vs political risk.
tweet

The Transcript
RT @TheTranscript_: AMD leads tomorrow's earnings:

Before Open: $PYPL $PEP $ETN $PFE $MRK $GLXY $HUBB $FUBO $ADM $MPLX

After Close: $AMD $SMCI $CMG $ENPH $LITE $AMGN $TTWO $PRU $DOX $CB $EA $AMCR $EMR https://t.co/SjKoxQY6Y6
tweet

Moon Dev
clawbot for trading

today we are doing day 2 of clawbot for trading

day 1 was crazy, she found some super profitable strategies using all my quant code

join here if there is still a ticket: https://t.co/JbJdIbW2p9

moon
tweet

Illiquid
More upside for $ter even after +25% move?
https://t.co/cosgBaKJqq
tweet

Jukan
Don’t get me wrong — I’m not “a person familiar with the matter.” I simply responded to the WSJ’s request for comment. https://t.co/d2TGLLfGuw
tweet

God of Prompt
RT @rryssf_: neurips 2025 gave its best paper award to research proving diffusion models have a "built-in alarm clock" that delays memorization.

the headlines say this solves the overparameterization paradox.

i read the whole paper. the limitations section tells a different story.

here's what the discourse is missing:
tweet

Jukan
RT @Citrini7: Okay, fine, we will make the Semis Memo monthly instead of quarterly. Please stop dm-ing me asking for this!

From here until the AI bubble pops or semis are cyclical again (just kidding, we all know that’s impossible now), we’ll be doing a quarterly Semis Memo.

Our first installment is a banger, focused on memory test, semicap subsystems & more.

https://t.co/wpXXiZTMu2 - Citrini
tweet

Dimitry Nakhla | Babylon Capital®
RT @DimitryNakhla: Dev Kantesaria is famously known for having very little trading activity — and an exceptional long-term track record.

When asked what his “secret sauce” is to staying away from the sell button, his answer was simple:

Focus on 𝐛𝐮𝐬𝐢𝐧𝐞𝐬𝐬 𝐪𝐮𝐚𝐥𝐢𝐭𝐲.

“It’s a much easier way to make money, it’s more tax efficient, and it’s a more predictable way to make money. So the lack of turnover is really nothing other than a natural outcome of how we think about business quality.”
___

Dev credits the Warren Buffett and Charlie Munger playbook — avoid short-term trading (speculation) and instead own companies that can compound intrinsic value year after year for decades.

This ties closely to Chris Hohn’s emphasis on long-termism.

𝘞𝘩𝘦𝘯 𝘺𝘰𝘶 𝘢𝘯𝘤𝘩𝘰𝘳 𝘰𝘯 𝘲𝘶𝘢𝘭𝘪𝘵𝘺 𝘢𝘯𝘥 𝘥𝘶𝘳𝘢𝘣𝘪𝘭𝘪𝘵𝘺, 𝘥𝘰𝘪𝘯𝘨 𝘯𝘰𝘵𝘩𝘪𝘯𝘨 𝘣𝘦𝘤𝘰𝘮𝘦𝘴 𝘵𝘩𝘦 𝘥𝘦𝘧𝘢𝘶𝘭𝘵 — 𝘢𝘯𝘥 𝘰𝘧𝘵𝘦𝘯 𝘵𝘩𝘦 𝘮𝘰𝘴𝘵 𝘱𝘳𝘰𝘧𝘪𝘵𝘢𝘣𝘭𝘦 𝘢𝘤𝘵𝘪𝘰𝘯.

Video: Good Investing Talks Podcast (10/27/2022)
tweet