Nathan Cooper (@ncooper57) 's Twitter Profile
Nathan Cooper

@ncooper57

The world can be ugly and cruel to the most innocent. Consider donating to help children suffering from one of the worst things: thorn.org/donate

ID: 1193330486359273475

linkhttps://nathancooper.io calendar_today10-11-2019 00:52:44

7,7K Tweet

1,1K Followers

857 Following

Google DeepMind (@googledeepmind) 's Twitter Profile Photo

We tested our controllers in hardware on the real LIGO system. Our results show that Deep Loop Shaping: 🔹controls noise up to 30-100 times better than existing controllers. 🔹can eliminate the most unstable, difficult feedback loop as a meaningful source of noise on LIGO for

We tested our controllers in hardware on the real LIGO system.

Our results show that Deep Loop Shaping:
🔹controls noise up to 30-100 times better than existing controllers.
🔹can eliminate the most unstable, difficult feedback loop as a meaningful source of noise on LIGO for
Jeremy Howard (@jeremyphoward) 's Twitter Profile Photo

Guess who was the 1st to point out that Adam can be used for pretty much everything? (Answer: it was fast.ai back in 2018 -- Sylvain Gugger's first research project in fact!) fast.ai/posts/2018-07-…

Guess who was the 1st to point out that Adam can be used for pretty much everything?

(Answer: it was <a href="/fastdotai/">fast.ai</a> back in 2018 -- <a href="/GuggerSylvain/">Sylvain Gugger</a>'s first research project in fact!)
fast.ai/posts/2018-07-…
Sayash Kapoor (@sayashk) 's Twitter Profile Photo

AI as Normal Technology is often contrasted with AI 2027. Many readers have asked if AI evaluations could help settle the debate. Unfortunately, this is not straightforward. That's because the debate is not about differences in AI capability, which evaluations typically measure.

AI as Normal Technology is often contrasted with AI 2027. Many readers have asked if AI evaluations could help settle the debate.

Unfortunately, this is not straightforward. That's because the debate is not about differences in AI capability, which evaluations typically measure.
David Gwyer - ML/AI engineer, n8n, MCP, researcher (@dgwyer) 's Twitter Profile Photo

Hi friends. 👋 Unfortunately I am struggling financially right now. I'm still looking for ML/DL roles in Python, LLMs, data science, automation (n8n) workflows. If you know of any leads please let me know. 🙏 Some of my work: github.com/orgs/Exploring… github.com/dgwyer

Google AI Studio (@googleaistudio) 's Twitter Profile Photo

official 🍌 Nano Banana global hackathon! build a product, app or demo that shows off the "wow" factor of gemini-2.5-flash-image, our SOTA image model weekend hackathon starts Sep 5 5pm PT in collab with ElevenLabs and fal details ⬇️

official 🍌 <a href="/NanoBanana/">Nano Banana</a> global hackathon!

build a product, app or demo that shows off the "wow" factor of gemini-2.5-flash-image, our SOTA image model

weekend hackathon starts Sep 5 5pm PT in collab with <a href="/elevenlabsio/">ElevenLabs</a> and <a href="/FAL/">fal</a>

details ⬇️
Hamel Husain (@hamelhusain) 's Twitter Profile Photo

Recap of tech ragebait I’ve seen in the last 24 hours - Serious work doesn’t get done remotely - Evals are dead - A/B testing is dead - If you aren’t using an agent framework in favor of while loops it’s just hubris - You can replace the PyData ecosystem only if there was

Anthony Hobday (@hobdaydesign) 's Twitter Profile Photo

I think macOS has lost its way, so I've switched to PostHog's new OS as my daily driver. When an operating system doesn't have something as basic as Godzilla window management, you know it's time to switch.

Tanishq Mathew Abraham, Ph.D. (@iscienceluvr) 's Twitter Profile Photo

Excited to announce that Sophont has raised $9.22M in combined pre-seed+seed rounds! 🚀🔥 Led by Kindred Ventures, with participation from @delphi_ventures Upfront Ventures AICONIC VENTURES also @jeffdean, @logankilpatrick, clem 🤗 (via Factorial Capital), Lukas Biewald & others

Simon Willison (@simonw) 's Twitter Profile Photo

Getting LLMs to write prompts for themselves is one of those things that I didn't really trust a year ago - but the latest generation of models (Claude 4, GPT-5 in particular) have impressed me with their promoting abilities simonwillison.net/2025/Sep/14/mo…

Jeremy Howard (@jeremyphoward) 's Twitter Profile Photo

This sounds right to me. So… can we now all agree to stop saying "agent" and say "tool loop" instead? It's the same # of syllables, and is much clearer.

DHH (@dhh) 's Twitter Profile Photo

"It takes ten years for the culture of a great company to fall apart once the CEO seat is given to someone without an engineering or product background. That's been the story of Boeing, Intel, and now Apple." world.hey.com/dhh/the-great-…

Yann LeCun (@ylecun) 's Twitter Profile Photo

Code World Model: producing code by imagining the effect of executing instructions and planning instructions that produce the desired effect.