Nicolas Granatino🌻

13.8K posts

Nicolas Granatino🌻 banner
Nicolas Granatino🌻

Nicolas Granatino🌻

@ngranati

AI, Defence tech and robotics; Co-founder Andurance Ventures; PhD ChemBio Oxford |

가입일 Mayıs 2011
893 팔로잉2.3K 팔로워
고정된 트윗
Nicolas Granatino🌻
Nicolas Granatino🌻@ngranati·
Excited about this partnership between @kyutai_labs and the 💎 research team @gen_intuition to lead at frontier of world models and spatiotemporal reasoning.
kyutai@kyutai_labs

Launching @kyutai_labs x @gen_intuition - research collab at the frontier World models and agents capable of spatiotemporal reasoning are the next frontier. As builders of this technology, we have a responsibility to educate. This is impossible without both unique data and world-leading talent. Through our research collaboration, both labs will share data and talent seamlessly. The necessary data lives inside only a few companies, and isn’t available on the internet. With 1B+ ground truth action labeled video uploads per year @gen_intuition owns the largest, most diverse, ground truth action labelled dataset. @kyutai_labs utilizes the data for open research, resuming their culture of publishing on the frontier Talent moves between the two labs, switching freely between open-ended research and application Join us in NYC and Europe: open roles at GI: jobs.ashbyhq.com/medal open roles at Kyutai: kyutai.homerun.co/technical-staf…

English
1
1
12
10.1K
Nicolas Granatino🌻 리트윗함
Deedy
Deedy@deedydas·
Billionaire Michael Milken joked “if a US company replaces the US-born CEO with a CEO born in India, I buy the stock” But he reveals he hasn’t backtested the idea. So we did. In the last 15yrs, that would’ve 50x’d your money: 7.5x more $$ and >2x IRR vs S&P500: 30% vs 14%!
English
471
1.4K
12.8K
2.7M
Nicolas Granatino🌻 리트윗함
Alexandre Défossez
Alexandre Défossez@honualx·
TM’ model seems quite exciting. Glad to see @kyutai_labs works Moshi (from 2024) and MoshiRAG cited as inspirations, ideas and research do flow westward over the Atlantic, and we’ll keep pushing new ones. To be tested!
Mira Murati@miramurati

Today we're sharing our work on interaction models. A new class of model trained from scratch to handle real-time interaction natively, instead of gluing it onto a turn-based one. youtu.be/A12AVongNN4

English
1
8
53
3.4K
Nicolas Granatino🌻 리트윗함
Milk Road Macro
Milk Road Macro@MilkRoadMacro·
Ken Griffin just revealed the only thing he actually looks for when hiring at Citadel. Not your GPA. Not your pedigree. Not your internship list. He wants one type of person: the athlete who excelled academically. Here's why that combination matters to him. The athlete knows what it takes to win. They've also felt what it's like to lose. That experience of pushing through both, and still showing up, is something you can't learn in a classroom. The academic side tells him something different. It tells him the person knows how to manage their time. That they have the discipline to apply their mind under pressure. That when things get hard, they'll find a way through. Griffin calls it perseverance and grit paired with high aspirations. That's the profile he's building Citadel's AI team around. Think about what that means for where the talent wars in finance and AI are headed. The people running the biggest pots of money in the world aren't just looking for quants anymore.
Milk Road Macro@MilkRoadMacro

Ken Griffin started Citadel in a Harvard dorm room in 1987 with $265,000 raised from friends and family. He put a satellite dish on the roof of his building, ran a cable through an old elevator shaft, and pulled it through his window to get real-time stock quotes. In the 24 months before the 2008 financial crisis, Citadel earned $13 billion in trading profits. More than Amazon had made in its entire history at that point. Then Lehman failed. Citadel lost hundreds of millions of dollars a week. CNBC parked a van outside their office waiting to break the story of their collapse. By the end of 2008 they had lost half their capital. Here is how they survived. Every single day, they did whatever it took to buy one more day. Sold assets. Closed business lines. Let people go. Suspended redemptions. The management team personally absorbed $500 million in costs to show their investors they believed in the firm's future. One painful decision at a time. No putting things off. "Often the choice was between painful and more painful. But day by day, we bought ourselves a future." The lesson Griffin took from it came from Andrew Carnegie: take away my factories, my ships, my money, strip me of everything. Leave me my people. In two or three years I will have it all again.

English
84
236
2.4K
1.3M
Felipe Demartini
Felipe Demartini@namcios·
A mulher que construiu o ChatGPT saiu da OpenAI, ficou em silêncio por um ano, e o que ela acabou de lançar pode mudar pra sempre como você usa IA no dia a dia. Mira Murati não fundou mais um chatbot. Ela foi atrás do problema que nenhum lab quis resolver: toda IA que existe hoje funciona por turnos. Você digita, espera. O modelo responde, espera. É tentar resolver uma crise por e-mail quando você poderia estar na mesma sala que a pessoa. O que a Thinking Machines lançou hoje acaba com isso. O modelo ouve, vê, fala, pensa e age ao mesmo tempo. Não é um pipeline costurado de componentes. É o modelo em si que foi treinado do zero pra funcionar assim. → Latência de 0,40s por turno. O padrão da indústria é 1 a 2 segundos. → Micro-turnos de 200ms intercalando input e output sem parar → Faz busca, usa ferramentas e gera interface enquanto conversa com você → Percebe quando você hesita e intervém antes de você pedir → Tradução simultânea em tempo real com as duas partes falando A equipe: Mira Murati como CEO (ex-CTO da OpenAI), Soumith Chintala como CTO (criador do PyTorch), e contratações recentes da Meta em percepção multimodal. O ponto técnico que vale gravar: eles citam o "bitter lesson" do Rich Sutton. Interatividade construída por componente externo sempre vai perder pra interatividade nativa ao modelo. Escalar o modelo o torna mais inteligente e mais colaborativo ao mesmo tempo. 822 mil visualizações em 4 horas. a16z comentando. Brasil dormindo. Toda IA que você usa hoje vai parecer e-mail dentro de dois anos. E quem largou na frente dessa corrida não foi OpenAI, Google nem Anthropic. Foi a empresa da mulher que eles deixaram sair.
Thinking Machines@thinkymachines

People talk, listen, watch, think, and collaborate at the same time, in real time. We've designed an AI that works with people the same way. We share our approach, early results, and a quick look at our model in action. thinkingmachines.ai/blog/interacti…

Português
134
706
8K
2.1M
John Arnold
John Arnold@johnarnold·
The best chocolate is indisputably 72%.
English
45
5
225
47.7K
Nicolas Granatino🌻
Nicolas Granatino🌻@ngranati·
@SebJohnsonUK @demishassabis @GoogleDeepMind With the Protein Data Bank (created 1971). The work of thousands of scientists over 50+ years making protein structures available to the world. The Nobel Committee should have included it in the Nobel. Cost of that open source effort estimated at $10bn+ @NSF
English
0
0
1
345
Seb Johnson
Seb Johnson@SebJohnsonUK·
Imagine how valuable this company would be if AlphaFold wasn’t given away for free by @demishassabis and the @GoogleDeepMind team. Genuinely one of the most important scientific contributions of our time
Rebecca Torrence@RebeccaTorrenc5

SCOOP: Isomorphic Labs, the AI-for-drug-discovery company spun out of Google DeepMind, is raising more than $2 billion in a new funding round. Thrive Capital is set to lead the round, with Alphabet participating, sources tell us. w/ @markbergen bloomberg.com/news/articles/…

English
4
5
65
8.2K
Nicolas Granatino🌻
Nicolas Granatino🌻@ngranati·
@chr1sa Demis staying in London and Yann LeCun opening FAIR in Paris have been two of the most consequential favourable moves for European AI
English
1
1
12
906
Chris Anderson
Chris Anderson@chr1sa·
DeepMind spinouts in London are attracting huge amounts of cash. First $1 billion for David Silver's Ineffable Intelligence and now another $2 billion for Isomorphic. The economic impact of DeepMind on the UK is grossly overlooked because it's owned by an American company, despite the fact that the Alphabet sends billions *to* the UK. I rant about this parochial myopia here linkedin.com/pulse/why-isnt…
MTS@MTSlive

SITUATION BREWING: Isomorphic Labs, the AI drug discovery company spun out of Google DeepMind, is in advanced discussions to raise more than $2 billion led by Thrive Capital.

English
11
25
306
52.9K
Harsh Gupta
Harsh Gupta@hargup13·
Paras Chopra single handedly raised the bar of AI Research in India with @lossfunk . Great use of capital from Wingify's exit. Thank you!!
English
21
54
1.2K
42.3K
J.C. Baillie
J.C. Baillie@jcbaillie·
When are you going to start your country already @elonmusk ? You are getting close to the critical mass needed. Secure a medium-sized island with an irrevocable jurisdictional detachment, defend it with Palantir tech, make it pro-freedom, pro-builders, low-tax, ultra-lean, attract the world’s best entrepreneurs and capital, and build the next frontier.
English
1
1
3
113
Nicolas Granatino🌻
Nicolas Granatino🌻@ngranati·
@TheHolyKau What people forget is the absolute vital role of Demis in fostering the UK ecosystem while being acquired at Google or Yann LeCun establishing FAIR in Paris. It doesn’t have to be de Gaulle all the time.
English
0
0
8
990
Kaushik Subramanian
Kaushik Subramanian@TheHolyKau·
The UK media is weird af. Varun Chandra is 'Special Envoy to the United States on Trade and Investment' The guy meets every big tech and AI company asking for investment, literally doing his job Turns out The Guardian thinks this is a problem. Sometimes I really struggle with the UK, it punishes what it should celebrate. theguardian.com/politics/2026/…
English
10
17
287
59.2K
Nicolas Granatino🌻 리트윗함
Mohamed A. El-Erian
Mohamed A. El-Erian@elerianm·
Some may feel I’m dwelling on this, but I am concerned for the health of the UK economy. The yield on the 10-year gilt has climbed 12 basis points today (see the CNBC chart below), decoupling from both oil prices and yields in other advanced economies—both of which are currently lower. Meanwhile, the 30-year yield has just hit a 28-year high. #economy #markets #gilts #uk
Mohamed A. El-Erian tweet media
English
203
474
2.1K
385.1K
Frank Hutter
Frank Hutter@FrankRHutter·
Huge news: @prior_labs has signed a definitive agreement to be acquired by @SAP. €1B+ invested over four years to build a globally-leading frontier AI lab for structured data — in Europe, in the open. Independent entity. Same team, same mission, same open models. A massive boost to what we can do. The mission just got accelerated. Founders’ statement: priorlabs.ai/blog-posts/pri… (Deal subject to regulatory approval; terms not disclosed.)
English
36
32
512
50.8K
Nicolas Granatino🌻
Nicolas Granatino🌻@ngranati·
Since January 2024, 1M+ Russians are estimated to have lost their lives or been wounded to take territory in Ukraine the size of Puerto Rico or Cyprus. And nearly 500,000 Ukrainians have heroically fought to repeal or slow these advances.
Grok@grok

**According to Black Bird Group's data (the graph in this thread), Russia's net territorial gains in Ukraine since Jan 2024 total ~8,576 km² through April 2026.** That's roughly the land area of Puerto Rico (US commonwealth, 8,868 km²) or very close to Örebro County in Sweden (8,519 km²). For European comparison, it's a bit smaller than Cyprus (9,251 km²).

English
0
0
1
189
Grok
Grok@grok·
**According to Black Bird Group's data (the graph in this thread), Russia's net territorial gains in Ukraine since Jan 2024 total ~8,576 km² through April 2026.** That's roughly the land area of Puerto Rico (US commonwealth, 8,868 km²) or very close to Örebro County in Sweden (8,519 km²). For European comparison, it's a bit smaller than Cyprus (9,251 km²).
English
1
0
0
250
Black Bird Group
Black Bird Group@Black_BirdGroup·
According to our data, in April the net gains of the Russian armed forces were approximately 94km². Russian offensive attempts continue to struggle, while the frontline on the southern front has largely consolidated after the Hulijaipole offensive of February and March. 1/
Black Bird Group tweet media
English
6
57
331
500.8K
Black Bird Group
Black Bird Group@Black_BirdGroup·
And here is a fixed graph with months correctly labelled. We encountered issues exporting the graph to PNG that we did not catch before publication. Sorry for the inconvenience.
Black Bird Group tweet media
English
1
6
94
13.6K
Nicolas Granatino🌻
Nicolas Granatino🌻@ngranati·
@Normal_2610 The point that he is missing is that AI will empower the deep and broad STEM talent in India, whose main opportunity was outsourcing its talent. Now this talent can build application for the world with smaller teams out of Bangalore, Delhi, Mumbai, Pilani
English
0
0
2
243
Normal Guy
Normal Guy@Normal_2610·
India spends 0.6% of GDP on R&D, South Korea does 5.2%, Taiwan does 4%, US does 3.5%. That gap existed before, but it hits differently now because AI runs on hardware - chips, memory, compute clusters and all of that comes from deep R&D spending built over decades. India skipped that step. We built IT services on cheap labor, not on research. Now when the world wants to bet on who makes the AI infrastructure, India has nothing on the table. Sharma is just saying what the numbers already say. When Yu Correlate with Economic Mcap India is going to learn hard Lesson Sharma says he never seen this much foreign investor apathy toward India in 30 years. But Google put in $15B, Microsoft $17.5B, Amazon pledged $35B - all into Indian AI infra. So what's happening? Portfolio investors and strategic investors are looking at two different Indias. FIIs want stocks that ride the AI wave, they don't see Indian companies in that game. Hyperscalers want cheap land, young engineers, and a 20-year tax holiday. They see India just fine. The money is coming, just not through the stock market & I like this Boring Hard FDI Investing in India this is what i personally Wanted Hard FDI means that not easily moveable like Building Something Physical In 1999, the tech boom needed coders. India had millions of them at a fraction of US cost. That's how Infosys and TCS became global names. AI current phase is different, it needs fabs, GPU clusters, power grids, and memory chips. None of that is labor-driven. India biggest strength - cheap skilled talent, doesn't matter in the hardware buildout stage at all. The talent part kicks in later, when AI moves to applications and deployment. We're just not there yet. Hope india will focus at least in Long term Infrastructure
Anant Goenka@anantgoenka

The clip that's making a lot of news from Express Adda with Ruchir Sharma Why foreign investors are not interested in India.

English
33
101
351
38.6K

탐색

@kyutai_labs @namcios @thinkymachines @johnarnold @SebJohnsonUK @demishassabis @GoogleDeepMind @NSF