What the smart people are recommending. 7867 books, tools, and products endorsed by the thinkers absorb.md tracks. Ranked by how many times each has been recommended across compiled podcasts, papers, posts, and tweets.
βa has a separate marketplace called a horizon that works for permission assets... companies like a are basically building versions of a in the case of a collab horizon that are already designed to takβ¦β
βyou can go to rw.xyz XYZ. You can see there's six different versions of tokenized coin which none of them is actually equity on Coinbase.β
βTo help address this gap, we introduce COMPOSITE-STEM, a benchmark of 70 expert-written tasks in physics, biology, chemistry, and mathematics, curated by doctoral-level researchers... All tasks are opβ¦β
βThere's this webpage called the Risk List, uh it's uh spelled uh with a Q instead of a C, uh by this company called Project 11, where they have this uh you know, this dashboard that gives you a live vβ¦β
βI wrote a short story about this called Trillion Dollar Salvage. Um I recommend you guys read it.β
βI did see a very cool website called PQ Ethereum.org or whatever.β
βThe Google uh Dan Benet Justin Drake paper was pretty notable in that they didn't actually publish the circuit. They publish a ZK proof of the circuit.β
βthere's this ticker, this token called Y LDS. It's the first ever SEC registered yieldbearing stable coin. You buy that token, you lend it into loans figure pools. YLDS earns itself 3.8% as a base.β
βmatching the algorithm of Karger and Levine for undirected graphs (SICOMP 2015)β
βThere is a link we'll include in the show notes. If you'd like to donate to the Roman Storm cause, please do so. You need some legal support and those funds will help.β
βIf you are holding ETH A in your brokerage, you might want to consider transitioning to ETHB.β
βI'm mostly outsourcing my brain on what's going to happen in the cycle to Michael Nato. I think he's got a pretty good pulse on it. He's been right consistently.β
βSurprisingly, we can black box the dynamic data structure from the recent interior point method-based flow algorithm of van den Brand et al. (FOCS 2024).β
βIf you have a DAP that can profit from this, be in touch with us. The website is ez.io. If you're an L2, same. If you're a user and you want to feedback on this or you just want to look at our GitHub,β¦β
βmorphing circuits introduced in Refs. [1-3]β
βMeta training and inference accelerator; It's Meta's first in-house silicon. It was designed solely for recommendation models in mind.β
βWe believe our Research Supercluster is one of the fastest AI supercomputers in the world. We have 16,000 GPUs interconnected with 1.6 Terabits per second of InfiniBand network.β
βI started my career at AMD it was an amazing place to work I learned so much during my time thereβ
βmy next three jobs were actually in startups where I had to get out of my comfort zone do different kinds of jobs...the scrappiness actually really really benefited meβ
βHe and Hua [Calc. Var. PDE, 2022] showed that $Ξ»_2(T) \leq 2/D$ for any tree of diameter $D$, with the even-diameter equality case fully characterized.β
βto do just that, Schollet created a special test. It's called the abstraction and reasoning corpus or arc for short. And the story behind this benchmark is fascinatingβ
βwe're going to focus on the work of a researcher named Francois Cholay. His core argument is that the way the industry defines artificial general intelligence or AGI is fundamentally missing the entirβ¦β
βI think it it's worth reading because it kind of goes through the um you know, the kind of the history of organizations and how almost like I don't know if they mentioned like the joint stock corporatβ¦β
βmorphing circuits introduced in Refs. [1-3]β
βYeah Quen 36 looks like a really good model. I haven't had a chance to play around with it yet. I definitely will.β
βGo check him out. If you want to learn everything about hardware, he has uh has some awesome takes.β
βmorphing circuits introduced in Refs. [1-3]β
βIt's these kinds of insights that make Mottz awesome to watch and uh I'm going to be looking out for the anthropic uh announcement that they've acquired Mott.β
βWayne super interesting what you're building. I uh I'm very happy that you came on and and explained it.β
βGo subscribe. Go like this. Uh, follow us on Twitter at ForwardFuture, forwardfuture.ai for our incredible newsletter daily.β
βSAM audio is a cutting-edge audio model with applications for musicians, audio engineers, video creators, and tinkerers of all skill levels. Learn more about SAM Audio today.β
βhe recently put an article out that you may have seen uh called from hierarchy to intelligence and it's a manifesto on how to rethink basically rethink from the first principles how an organization woβ¦β
βThe waiting list for Henry is now open at meethenry.ai.β
βwe ship something called missions, which lets Droid operate sort of a team of agents for...the longest mission we saw run was 14 days straight of continuous effort from...hundreds of agents along the β¦β
βwe have our like uh auto wiki uh product which basically builds a continuous think of it as like internal engineering system overview over the different systems involvedβ
βWhat are you most excited about for AI Council this year?...everyone's coming in with a totally different perspective on agentic developmentβ
βIt comes from a 1988 book mind children by Hans Moravec who was and is a robotics researcher.β
βThe one that is actually the most cost-effective if you want to use Google Flow is going to be Google AI Pro. This is the tier that is actually at a reasonable price, around $20 a month, and you get aβ¦β
βIf you're on the Google AI Ultra tab, essentially, you do get more bang for your buck, but it is of course more expensive at around $200 to 250 pounds per month. I would say only get this if you're goβ¦β
βNow that this essentially covers the bases of your images, additionally, don't forget that if you want even more control, what you can do is use the draw feature.β
βFor example, the Juicero was the juice machine that famously failed after being introduced at a ridiculous price. It was destined to flop because the machine was extremely overengineered.β
βIn my book, AI snake oil with Tayash Kapoor, we predict that the same thing will happen with AI.β
βSteve Newman, who's been doing some great writing about AI on his Substack newsletter, gives a good example of this in the context of his being disappointed over GPT5.β
βA new report from the UK's AI safety institute talked to a bunch of AI experts to identify a list of things that today's AI fundamentally can't do and which of those things are probably necessary befoβ¦β
βIf you want to understand mythos, you need to understand how Anthropic actually organizes their models. So at the bottom you've got Haiku, which is cheap and lightweight. Then you've got Sonic, which β¦β
βSo, right now, you can see that I am in the App Store and essentially all you'll need to do is just download this. There is no weight list. There is no specific things you need. You don't need a develβ¦β
βIf your Android phone has 8 GB of RAM or more and was released in the last few years, you can probably run the Gemma 4 models. If it has 12 GB of RAM, you can probably handle the models too.β
βAnd of course, the smaller the model is, you can see down here you've got Gemma 3 1 billion parameters that is going to be a model that is going to be easily run on the older devices that we mentionedβ¦β
βIt's actually possible to calculate the shape of these bubbles as a function of temperature. Look up the wolf shape if you want to learn more.β
βSomething slightly more accessible, which you might have heard of before, is the Ising model. This is a model of a magnet which consists of a bunch of tiny magnetic elements arranged in a grid, each oβ¦β
βThis is called the XY model. If you are curious and want to learn more, all of these models can also be considered in three dimensions, or even higher, or on other strange non euclidean geometries.β
βIf you are using certain open- source models, maybe they may not have the same performance as Opus 4.5 or GPT 5.4 or GPT 5.3 codecs. And you should take that into account because remember every singleβ¦β
βIn this paper, we introduce PROMPTEVALS, a dataset of 2087 LLM pipeline prompts with 12623 corresponding assertion criteria, sourced from developers using our open-source LLM pipeline tools.β
βWe present gpt-oss-120b and gpt-oss-20b, two open-weight reasoning models that push the frontier of accuracy and inference cost.β
βTogether with the results of He-Hua and Lin-Zhao [Bull. Lond. Math. Soc., 2025] for even diameters and diameter three, this completes the geometric classification for every diameter.β
βThe inverse variational viewpoint may be regarded as a boundary analogue of the classical distance-matrix formalism for trees initiated by Graham and LovΓ‘sz [Adv. Math., 1978].β
βWe advocate for the alternative construction of the effective action and potential based on the Vilkovisky-DeWitt approachβ
βWe propose CRISP, which instead jointly ranks all cited papers within a citing paper using large language models (LLMs).β
β# MycoBenefitsβ’ Focus*: A Natural Supplement for Focus, Clarity, and Cognitive Support*β
βNew Research Highlights Distinct Support for Immune-Balancing Activity in Lionβs Mane Mushroom Mycelium as Compared to Fruiting Body Extractsβ
βBlog post by Paul Stametsβ
β_Blog post by Laura Martin_β
β125 students and 100 volunteers put engineering in action for National Robotics Weekβ
βthere's the institute for progress β«in DC that I think is doing some excellent work hereβ
βShe has a company called Imagi and they help kid that educate uh about computer science AI and so on.β
βI learned about this organization Burtzorg in in uh the Netherlands. That's probably the most radical because they have 16,000 employees and two managers. Interesting. Okay. What's it How do you spellβ¦β
βThis is the perspective of Viktor BlΓ₯sjΓΆ, who hosts the excellent podcast Opinionated History of Mathematics, which features an 18 episode season about Euclid's elements and the history of geometry.β
βI just started a YouTube channel on trying to advance the public understanding of science called for the love of science but I don't really know how to use YouTube so we just are posting videos uh youβ¦β
βClaire Vo is the host of our sister podcast, βHow I AI,β a former product executive and engineer, and founder of an AI startup called ChatPRD. Claire now runs her business, podcast, and family life wiβ¦β
βIf you want something that's sort of gentle... a gentle bedtime podcast, I'd say The History of English is quite a nice oneβ
βOn Investment Management Operations, Scott MacDonald speaks with Henry Ward of Carta on building the ERP for private markets, navigating pivots, and where AI could reshape SaaS.β
βHey, can your week feels like a lifeter of the Visual Learners channel where we decode the future of how we create.β
βIf you found this deep dive into the design wars of 2026, as fascinating as I did, do me a favor, hit that like button and subscribe to VL Extras.β
βif you want to see the best of those click over to the initialized capital Channel and hit subscribe there Link in the descriptionβ
βSam is the author of five New York Times bestsellers as well as the host of a top 10 podcast making sense where he explores some of the most important questions about the human mind society and currenβ¦β
βHe has a podcast called All the Hacks which he has used to explore some really fascinating stuff. If you're a points nerd or like travel, it's a good one.β
βI think Christa Tippet, great podcaster by the way, but Christa Tippet, one of the OGs on Being, I believe, is her podcast.β
βAmericans podcast that we just dropped, the third one. You guys should watch that. We are recording the fourth one this coming week, so that will be dropping also.β
βHow Code-Switching Explains The Worldβ
βThe follow up is live now on the Spectral Collective channel. There we'll look at a simplified version of the liquid vapor model, which we can essentially completely understand using basic mathematicsβ¦β
βAccess to my Content Catalyst custom-built AI tool that help you write content that actually sells.β
βHow to Be Cultured https://www.nytimes.com/interactive/2026/04/17/t-magazine/culture-guides-film-art-food-literature.html?smid=tw-shareβ
βI also have a weekly newsletter. So, even though I might be on a little break from the podcast, you can actually go to my weekly newsletter to get insights on what I think about the future of the creaβ¦β
βDr. Lichtenstein says the process for making these new guidelines was to dispense with politics and stick to the evidence. Well, I think there's no agenda behind these guidelines that they're evidenceβ¦β
βAnnouncing the Cohere Partner Programβ
βTo hear more conversations from our Opinion colleagues, you can subscribe to the Opinions wherever you get your podcasts.β
βIf, Why, and How Founders Should Hire a βProfessionalβ CEOβ
β# The Singularity Is Not Coming _Blog post by FranΓ§ois Chollet_β
βin my luggage, I always always always 100% carry with me Valona unsweetened cocoa powder.β
β# Productivity _Blog post by Sam Altman_β
βI encourage you to check it out. We will link it in the show notes.β
βFor more insights on the survey, check that out at the information.com.β
βThis free 3-day experience helps you rewire your relationship with money...β
βPay us and you'll wait less. And it's pretty compelling in this age of long lines everywhere.β
βyou mentioned later in this wired article that you stumbled upon the single algorithm Theory popularized by Jeff Hawkinsβ
βthe prevailing opinion was that human intelligence derived from thousands of simple agents working in concert this is what mit's Marvin Minsky called The Society of mindβ
βthere's also a recipe Pioneer by one of my friends Eric Bron and his collaborators sanit called a task-based analysis of jobs that I found to be very usefulβ
βBlog post by Naval Ravikantβ
βyou could save a lot of money by working with GPT 40 mini instead of GPT 40β
βI'm hosting a virtual halfday conference on a gentic AI tomorrow it'll be interactive practical and it'll feature some of the most influential people in the AI agent space as speakersβ