illeit @_illeit

Goldman Sachs CEO: Tokenized equities is something we’re thinking a lot about David Solomon asks Coinbase CEO Brian Armstrong how he sees tokenized equities unfolding. Brian responds: “A few years ago, people were talking about stablecoins, and they were saying, ‘Why do we need a digital dollar? We kind of can already make digital payments… What’s the point of that?’ It turned out to be a massive market. There’s high demand for the dollar in all these countries around the world, and a lot of people can’t get access to dollar-denominated accounts — they live in a high-inflation country like Turkey or Argentina or Nigeria. So that was one piece. And the other was it just reduced friction in terms of all kinds of payments people wanted to make in crypto for trading and B2B payments. There’s something like $30 trillion of stablecoin payment volume in the last year.” Brian believes tokenized equities will follow a similar path: “We don’t know exactly how it’s going to play out, but if you simply store a share of a company with a traditional custodian and issue a token equivalent of it on-chain, what does that enable? Similar to stablecoins, there’s an international component to this. There’s lots of people in the world who would love to buy Tesla or Nvidia. If you’re wealthy in Argentina, you can probably get a brokerage account open somewhere to trade that kind of stuff, but the vast majority of people cannot. There’s also this 24/7 trading aspect — people want to do that, and I think crypto is going to get there faster. There’s fractional shares, where you can trade a tiny piece of a share. There’s things like perpetual futures markets which have done really well in crypto — why shouldn’t that exist for securities as well? I think you’re going to see some things like this where crypto reduces friction and allows people to try new things. There might even be novel governance things which get created: let’s say you want to create a stock where short-term holders of your stock can’t vote — you can program that into a smart contract on chain and make that new type of token with those capabilities.” And it’s not just equities. Treasuries, private credit, real estate, and many other real world assets (RWAs) are being tokenized, with Ethereum is the preferred settlement rail for compliant institutional capital markets. More than 60% of all tokenized assets — over $200 billion — reside on Ethereum. Source: @GoldmanSachs (Dec 2025)

Look everyone... don't ever be concerned with my positions; be concerned with the risk you are taking. I've been trading for 43 years and have had only a few single-digit drawdown years. All of my trades are made from very low-risk entry points and always with a hard stop loss in place. I could turn out to be wrong and the market may have already made its low. If so, I'll get stopped out. That's the business. I'm wrong just as often as I'm right. The difference is that my risk is always defined and controlled. What matters is not being right all the time; what matters is that the risk taken relative to the potential reward, adjusted for batting average, is managed in a way that produces a profitable outcome over a large sample of trades. That's how I've approached the market throughout my entire career, and it's no different today. The distribution of gains and losses over time forms a profitable bell curve because risk always comes first, and risk is always managed in relation to reward.

Mark Minervini @markminervini

4 days ago

A few minutes ago I shorted the $QQQ - 🙏 Still short $SPY.

205 327 3K 3.6M 879

do you understand what Anthropic just admitted? their engineers haven’t written most of their own code since early 2026. 80% of the code merged into Anthropic’s codebase last month was written by Claude. let that sink in. the company building the most powerful AI in the world is already being built by that AI. and the numbers keep getting wilder: → engineers are shipping 8x more code than in 2024. not because they’re working harder. because Claude is doing most of it. → Claude’s success rate on open-ended coding problems hit 76% in May 2026. up 50 points in just 6 months. → one Anthropic employee said “it’s been 5 months since I last wrote any code myself.” → Claude Mythos Preview achieved 52x speedup on research optimization. a skilled human gets 4x in 4-8 hours. → the length of tasks AI can reliably complete is doubling every 4 months. and here’s the part that keeps me up at night: in April 2026, Claude-powered agents were given an open AI safety problem and left alone to solve it. two human researchers recovered 23% of the performance gap in a week. the agents recovered 97% in 800 hours. Anthropic calls this recursive self-improvement. AI building AI. getting better. building a better version. repeat. they say we’re not there yet. but they also say it could come sooner than most institutions are prepared for.

Anthropic @AnthropicAI

a week ago

Our internal data shows Claude is accelerating AI development—a possible path to recursive self-improvement, or AI autonomously building a more capable successor. It’s happening faster than we thought, and the implications deserve greater attention. anthropic.com/institute/recu…

2K 5K 29K 18.4M 15K

The thing is, it’s actually really good advice.

Jimmy Heaters @CathPoaster

2 weeks ago

new grads often ask me what they should be doing so they don't fall behind in the ai space. there's a lot, but its honestly super manageable. become intimate with model internals. proof based linear algebra. non-convex optimization. this is stuff you could've done in undergrad.

103 250 4K 738K 6K

Powerful and true. You should always listen to people you respect, but you have to make your own decisions. My parents often told me that I never listened. I told them that was not true. I always listened very carefully, but I always made my own decisions. You have to live your own life.

Race @multiplanet1

2 weeks ago

Elon Musk cried on national television when his childhood heroes called him a fraud. Neil Armstrong and Gene Cernan, the first and last men to walk on the moon, publicly testified against SpaceX. They said Musk was reckless. That private spaceflight was dangerous. That he was

374 1K 13K 1.5M 2K

C’è anche la versione da casa.

Make the Luce 30% wider and everyone would love it

Software went from desktop-first to mobile-first, now going to agent-first.

The new competition isn’t Humans vs AI. It’s Humans with AI vs everyone else.

Interesting.

Lenny Rachitsky @lennysan

3 weeks ago

My biggest takeaways from @danshipper: 1. The future of work will happen inside Codex or Claude Code. Instead of putting AI into your SaaS tool, you’ll use your SaaS tools inside your favorite AI agents' in-app browser. Dan spends all his time in Codex now—writing documents,

152 238 2K 761K 4K

Anthropic pays $750,000+ a year for engineers who know how to build LLMs from scratch. Stanford just released the exact lecture that teaches it - 1 hour 44 minutes, free, straight from CS229. Bookmark and watch it this weekend. It'll teach you more about how ChatGPT & Claude actually work than most people at top AI companies learn in their entire careers.

Rahul @sairahul1

3 weeks ago

x.com/i/article/2057…

66 418 2K 6.5M 11K

HOLY SMOKES! Andrej Karpathy warned about this a month ago and today he announced he's joining Anthropic. A month ago, Karpathy said openly that if you are outside a frontier lab, your judgment will inevitably start to drift. You lose touch with what is actually being built, how these systems work under the hood, and where the entire field is heading next. He said being inside one of the frontier labs doing really good work for some period of time might be the only way to stay genuinely connected to what is actually happening at the cutting edge. Today he acted on exactly that. Andrej Karpathy just announced he’s joining Anthropic’s pre-training team, placing him directly inside the most compute-intensive and technically demanding layer of building frontier AI models. Pre-training is where the large-scale compute runs happen, where the fundamental capabilities of a model are baked in at the deepest level and where the gap between frontier labs and everyone else is either won or lost permanently. He will also build and lead a new team focused on using Claude itself to accelerate pre-training research meaning Anthropic is now using its own models to help design and build the next generation of its models, closing the self-improvement loop that every major lab is racing to complete. The choice of Anthropic specifically is the signal worth paying attention to. Karpathy co-founded OpenAI, ran AI at Tesla for years, and spent the last two years as one of the most credible and widely followed independent voices in the entire field. He had every option available to him, OpenAI where he helped build the original research culture, Google DeepMind, xAI and he chose Anthropic. When the person who arguably understands AI pre-training better than almost anyone alive looks at the entire landscape and decides that Anthropic is where the most important work of the next few years will happen, that is not a career decision but rather a verdict on which lab is actually winning the research race right now.

Andrej Karpathy @karpathy

3 weeks ago

Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time.

8K 11K 150K 27.5M 14K

Karpathy's career moves are the single most accurate map of AI's center of gravity over the last decade. 2015: co-founds OpenAI when the frontier is pure research. 2017: leaves for Tesla when applied neural networks at scale become the hardest unsolved problem. Stays five years building Autopilot. 2023: returns to OpenAI during the GPT-4 sprint. Stays 12 months. 2024: launches Eureka Labs to teach the world how LLMs work. Now he's at Anthropic. The "get back to R&D" line tells you something specific. Karpathy spent two years on education. His YouTube lectures reach millions. Eureka Labs had real momentum. He walked away from a growing education business to join a company that went from $87 million in annual revenue to $30 billion in 28 months. When the person who teaches the world how neural networks work decides the opportunity cost of teaching is too high, the R&D window just entered a phase that won't stay open. He's pricing his own time against the frontier, and the frontier won. The career pattern is the real signal. Five years at Tesla. Twelve months at OpenAI the second time around. An OpenAI co-founder chose Anthropic. Draw whatever conclusions you want from that sequence.

Andrej Karpathy @karpathy

3 weeks ago

Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time.

8K 11K 150K 27.5M 14K

The hottest new programming language is English

Where will AI be in 1, 2 or 3 years?

The crypto market structure bill has PASSED the Senate Banking Committee with a bi-partisan vote! Historic day for crypto and for the future of digital assets in America. Grateful for the countless hours from lawmakers and staff to strengthen this legislation. Big improvement from where we were in January on rewards, tokenization, DeFi, and CFTC authority. I'm proud we stood up for our customers in that moment, and the bill is better because of it. Looking forward to a bipartisan law that cements the US as the world's crypto capital. Let's get CLARITY done.

How to become dangerously good at AI without wasting 1000+ hours. No useless tutorials. No fake AI gurus. No information overload. I spent weeks filtering the internet so you don’t have to. Here’s the ultimate AI learning stack for: • LLMs • AI Agents • MCP • Prompt Engineering • RAG • AI Engineering • Vector Databases 🧠 Videos LLM Introduction youtube.com/watch?v=zjkBMF… LLMs from Scratch youtube.com/watch?v=9vM4p9… Agentic AI Overview (Stanford) youtube.com/watch?v=kJLiOG… Building & Evaluating Agents youtube.com/watch?v=d5ElIX… Building Effective Agents youtube.com/watch?v=D7_ipD… Building Agents with MCP youtube.com/watch?v=kQmXtr… 🗂️ Repositories Microsoft AI Agents for Beginners github.com/microsoft/ai-a… Prompt Engineering Guide github.com/dair-ai/Prompt… Hands-On LLMs github.com/HandsOnLLM/Han… Made With ML madewithml.com LLM Course github.com/mlabonne/llm-c… 📚 Guides Google Agent Whitepaper kaggle.com/whitepaper-age… Building Effective Agents by Anthropic anthropic.com/engineering/bu… OpenAI Practical Guide to Agents platform.openai.com/docs/guides/ag… 📖 Books Building LLMs from Scratch manning.com/books/build-a-… The LLM Engineering Handbook oreilly.com/library/view/l… AI Engineering oreilly.com/library/view/a… 📄 Papers ReAct arxiv.org/abs/2210.03629 Toolformer arxiv.org/abs/2302.04761 Generative Agents arxiv.org/abs/2304.03442 🎓 Courses HuggingFace Agents Course huggingface.co/learn/agents-c… MCP with Anthropic anthropic.com/engineering Bookmark this. You’ll need it sooner than you think.

Before Elon Musk built rockets, he read every textbook on rocket propulsion ever written. Not summaries. Not blog posts. Not YouTube videos. The actual textbooks. Cover to cover. Multiple times. Orbital mechanics. Propulsion physics. Materials science. Structural engineering. Aerospace manufacturing. He did this while running two other companies. When he walked into meetings with aerospace engineers in the early days of SpaceX, they expected a tech billionaire who would nod along and sign checks. Instead he asked questions so specific they assumed he had a PhD in the field. He didn't. He had a library card and an obsession. This is the part of the Musk story that nobody wants to hear because it's not glamorous. The rockets are glamorous. The Tesla reveal is glamorous. The Mars vision is glamorous. Sitting alone in a room reading a 600 page textbook on orbital mechanics at midnight is not glamorous. But it's the actual foundation of everything else. Most people who want to "be like Musk" study his habits. His schedule. His management style. His tweets. They skip the reading because reading is boring and doesn't look good on social media. Musk reads 5 to 10 hours a day. He has done this since childhood. His brother Kimbal said he would read two books a day as a kid. Not children's books. Encyclopedias. Science textbooks. Science fiction. History. By the time he was 20 he had read more books than most people read in a lifetime. By 30 he had absorbed the foundational knowledge of four or five separate industries. By 40 he could hold his own with the world's top experts in rockets, batteries, AI, tunneling, neuroscience, and manufacturing. This is not genius. This is volume. The reason most experts can't do what Musk does is not intelligence. It's that they read deeply in one field and broadly in none. They know everything about their domain and nothing about any other. Musk reads deeply across six domains which means he sees connections that single domain experts can never see. Battery chemistry from Tesla applied to energy storage. Manufacturing techniques from automotive applied to rocket production. Software architecture from PayPal applied to vehicle autonomy. AI research from xAI applied to robotics. Each domain feeds every other domain because one person holds all the knowledge simultaneously. This is the real superpower. Not IQ. Not work ethic. Not risk tolerance. Cross domain knowledge accumulated through thousands of hours of reading that nobody sees and nobody talks about because it's not exciting content. I started reading seriously two years ago. Not scrolling articles. Reading actual books and technical material for hours. The difference in pattern recognition after 6 months was noticeable. After 12 months it was dramatic. Ideas from one domain started solving problems in completely different domains. The secret to Musk isn't in his tweets or his rockets or his schedule. It's in the thousands of books nobody saw him read before any of it existed. Reading is the most underpriced asset in the world. Unlimited knowledge, zero cost, available to anyone. Almost nobody uses it.