Podcasts about KV

Kvällspasset är ute och cyklar! Vi efterlyser lyssnarnas mest oväntade, roliga eller härliga berättelser från cykeln. Lyssna på alla avsnitt i Sveriges Radios app. Ett nyfiket och underhållande aktualitetsprogram med lyssnaren i fokus.Tobias BMX-tur tog tvärstopp när hunden Roj inte ville fortsätta, under en cykeltur fick Susanne fågelbajs i munnen och Hugo cyklade från Landskrona till Malmö för en konsert. Dessutom snackar vi med ”Taximami” Cathrine som berättar hur jobblivet är nu när alla andra är på semester!

Kvůli sinicím a bahnu klesá i množství kyslíku ve vodě a rybáři pravidelně řeší masivní úhyny ryb. Nejhoršímu mají zabránit nové provzdušňovače, které koupila obec. Nejúčinnější by ale bylo odbahnění.

I dagens avsnitt reser vi till piraternas guldålder omkring år 1715 – till en vit sandstrand i Karibien med palmer och turkost vatten! Kvällens saga "Louis piratäventyr" är önskad av Loui, 4 år från Vetlanda.Kapten Loui ger sig ut på de sju haven med hela familjen ombord. När de möter den läskiga Kapten Kroktand verkar det bli ett rejält sjöslag – tills lillebror Molle kryper fram med sin banan och frågar "Pirat vill banan?". Och då mjuknar den läskigaste piraten av alla. En varm saga om att även busar kan bli vänner.I studion gräver gänget upp en riktig skattkista – men inuti finns bara ett brev och en enda guldslant från piraten Svarte Samuel, som gav bort hela sin skatt till en fattig familj. Och Tobias blir bästis med papegojan Kapten Krax, som härmar allt han säger. Plus fem fakta om pirater – visste du att de hade demokrati ombord och delade rättvist på guldet?Stötta podden och få tillgång till nya sagor! Gå med i Magiska Godnattsagor-klubben!Skicka in förslag på kommande sagor via www.magiskagodnattsagor.seSökord: magiska godnattsagor, godnattsaga, barn, läggdags, podcast för barn, barnlitteratur, ai, godnatt, pirater, skattjakt, karibien, papegoja, vänskap, Vetlanda

Eliška Krausová odjela v létě 1968 na několik měsíců do Kolumbie zdokonalit se ve španělštině. Kvůli invazi vojsk Varšavského paktu tam ale neplánovaně zůstala a zemi neopustila ani po pádu Sovětského svazu. Jak nyní vypadá tamní politická situace? „Polovina Kolumbijců volila jednoho kandidáta a skoro stejná polovina toho druhého,“ shrnuje dění v pátečních Hovorech Českého rozhlasu Plus pedagožka na univerzitě v Bogotě.

Det är ju så pinsamt att göra bort sig! Kvällen bjuder på dråpliga händelser och vi undersöker varför vi egentligen tycker det är så jobbigt. Lyssna på alla avsnitt i Sveriges Radios app. Vissa gånger vill man bara försvinna, andra, landar man i ett gapskratt och ofta blir det en bra historia i efterhand. Karlavagnen handlar ikväll om att göra bort sig och känna att man bara vill sjunka under jorden när det går fel. Som när man snubblar på en trottoarkant, pruttar i fel miljö eller spiller på sin bordsgranne under en fin middag. Kvällens programledare Sofia Rågenklint kommer själv bjuda på några pinsamma ögonblick och har du också en historia när du gjorde bort dig rejält så ring oss på 020-22 10 30, mejla på karlavagnen@sverigesradio.se eller skriv till oss på Facebook och Instagram. Slussen öppnar kl 21:00 och programmet börjar 21.40

Eliška Krausová odjela v létě 1968 na několik měsíců do Kolumbie zdokonalit se ve španělštině. Kvůli invazi vojsk Varšavského paktu tam ale neplánovaně zůstala a zemi neopustila ani po pádu Sovětského svazu. Jak nyní vypadá tamní politická situace? „Polovina Kolumbijců volila jednoho kandidáta a skoro stejná polovina toho druhého,“ shrnuje dění v pátečních Hovorech Českého rozhlasu Plus pedagožka na univerzitě v Bogotě.Všechny díly podcastu Hovory můžete pohodlně poslouchat v mobilní aplikaci mujRozhlas pro Android a iOS nebo na webu mujRozhlas.cz.

Ján 5,1-9a 1Potom bol židovský sviatok a Ježiš vystupoval do Jeruzalema. 2V Jeruzaleme pri Ovčej bráne je rybník po hebrejsky nazývaný Betesda, s piatimi stĺporadiami. 3V nich ležalo množstvo chorých, slepcov, chromých, ochrnutých. 5Bol tam človek, chorý už tridsaťosem rokov. 6Keď ho Ježiš videl ležať a spoznal, že je už dlho chorý, opýtal sa ho: „Chceš vyzdravieť?“ 7Chorý mu odpovedal: „Pane, nemám človeka, čo by ma zaniesol do rybníka, keď sa zvíri voda; kým ta dôjdem sám, zostúpi predo mnou iný.“ 8Ježiš mu povedal: „Vstaň, vezmi si lôžko a choď!“ 9A ten človek hneď ozdravel, vzal si lôžko a chodil. Nemáš nikoho? Prichádza Ježiš. Koľkokrát sme o tom zapochybovali? Nemám nikoho; čas aj okolnosti hrajú proti mne; som stratený. Závisieť len od vlastných schopností, či od iného človeka alebo od toho, ako sa situácia vyvinie, sa v konečnom dôsledku končí beznádejou. Občas sa nám podarí niečo vyriešiť, ale sú veci, na ktoré nestačíme. Kvôli našej vnútornej chorobe – hriechu – sme všetci paralyzovaní. Možno to nevieme presne takto pomenovať, ale faktom je, že neustále hľadáme pomoc. Chytáme sa aj slamky, hoci je pri nás živý Boh. Živý, avšak nepovšimnutý. Je to paradox nášho bytia, pri ktorom by sme sa mali zastaviť. „Veríme“, ale nedôverujeme. A tak sa nám Pán prihovára vždy znovu a znovu – v spoločenstve aj osobne – aby sme si uvedomili, že o nás vie a záleží Mu na nás. Ako odpoveď stačí jeden malý krôčik z našej strany: priznať si svoju bezmocnosť. Sebe, no najmä Jemu. Tá modlitba v komôrke srdca nás môže zachrániť práve dnes. Veď Betezda – Dom milosrdenstva a súcitu – je tam, kde je s nami Pán Ježiš. Nie je pravda, že nemáš nikoho! Máš Toho, Ktorý ťa miluje plnosťou lásky. Modlitba: Pane, som bezmocný a je pre mňa také ťažké priznať si to, lebo bezmocnosť sa v tomto svete nenosí. Ďakujem, že Tebe to môžem bez obáv povedať, že Tebe môžem povedať všetko. Dávaj mi Svojho Ducha dôvery, aby som bol zachránený a žil vďaka tomu, že Ty si so mnou! Prosím nielen za seba, ale aj za mojich blízkych a blížnych. Amen. Pieseň: ES 490 Autor: Renáta Rényeiová Pozri z neba a hľaď zo Svojho svätého a slávneho príbytku! Kde je Tvoja horlivosť a odvaha? Izaiáš 63,15 Ježiš hovorí: „Nenechám vás ako siroty, prídem k vám.“ Ján 14,18 1.Jána 5,6-10 •  Modlíme sa za: Trenčianske Stankovce (PoS) Otázky na rozjímanie: Ako dnes žijem z istoty, že „Nemáš nikoho? Prichádza Ježiš“ — či som ochotný/á priznať si svoju bezmocnosť sebe, no najmä Jemu, alebo som stále plný/á toho, že som stratený, závisím len od vlastných schopností, alebo od iného človeka, a končím beznádejou? Kde v mojom živote som ako chorý, ktorý už tridsaťosem rokov leží a nemá človeka, čo by ho zaniesol do rybníka — či som ochotný/á uvedomiť, že o nás vie a záleží Mu na nás, a že Betezda — Dom milosrdenstva a súcitu — je tam, kde je s nami Pán Ježiš, alebo som stále plný/á toho, že neustále hľadáme pomoc a chytáme sa aj slamky, hoci je pri nás živý Boh, ale nepovšimnutý? Ako dnes žijem z výzvou „Máš Toho, Ktorý ťa miluje plnosťou lásky“ — či som ochotný/á mať Svojho Ducha dôvery, aby som bol zachránený a žil vďaka tomu, že Ty si so mnou, alebo som stále plný/á toho, že veríme, ale nedôverujeme, a preto sa nám Pán prihovára vždy znovu a znovu? Aplikácia do života: Dnes si spomeňte na jednu situáciu, kde ste boli ako chorý, ktorý už tridsaťosem rokov leží a nemá človeka, čo by ho zaniesol do rybníka (bezmocnosť, strata, závislost od vlastných schopností). Napíšte si modlitbu: „Pane, som bezmocný a je pre mňa také ťažké priznať si to, lebo bezmocnosť sa v tomto svete nenosí. Ďakujem, že Tebe to môžem bez obáv povedať, že Tebe môžem povedať všetko. Dávaj mi Svojho Ducha dôvery, aby som bol zachránený a žil vďaka tomu, že Ty si so mnou! Prosím nielen za seba, ale aj za mojich blízkych a blížnych!“ Potom urobte jeden krok priznania si svojej bezmocnosti: priznajte si svoju bezmocnosť sebe, no najmä Jemu, alebo modlite sa v komôrke srdca. Dnes som vďačný za tieto 3 veci: _________________________________ _________________________________ _________________________________ Viac o vďačnosti, čo to je, prečo je dôležité byť vďačný, ako praktizovať vďačnosť nájdeš na blogu

Landsskuet 2026 er overstået, og i denne udgave af Kødkvæg i Fokus får Anders og Henning besøg af Martin Lodahl Larsen, der løb med den helt store Interbreed-sejr. Martin fortæller om vejen til toppen, arbejdet med sin Dexter-besætning og livet som både kvæginseminør og passioneret avler. Samtidig ser værterne tilbage på årets Landsskue, de største højdepunkter og hvorfor netop dyrskuet fortsat er et af årets vigtigste samlingspunkter for dansk kødkvæg. I Kødkvæg i fokus sætter vi fokus på dansk kødkvæg – fra hverdagen i stalden til de store linjer i branchen. Serien tager fat på alt fra management, avl og kalvning til naturafgræsning, afsætning, økonomi og livet omkring kødkvæget. Med afsæt i praksis, erfaringer og faglig viden giver vi plads til både nørderi, holdninger og nuancer, så kødkvæget kan blive belyst fra flere vinkler. Kødkvæg i Fokus er støttet af Kvægafgiftsfonden.

En ung kvinde stiger ind i en blå og hvid Cadillac i den tro, at hun har mødt hjælpsomme fremmede, men turen gennem natten fører hende direkte til det gule hus i Glendale. Mens hendes skæbne folder sig ud bag lukkede døre, vokser frygten i Los Angeles, og navnet Kvæleren fra bakkerne begynder for alvor at sætte sig i byens bevidsthed. Samtidig dykker fortællingen ned i Angelo Buonos fortid, hans kvindehad og de mønstre af vold og kontrol, der længe før drabene har præget hans liv. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Summit Severoatlantické aliance v turecké Ankaře patřil kvůli kompetenčnímu sporu prezidenta a premiéra mezi asi ty vůbec nejsledovanější v Česku. Kvůli vnitropolitickým tahanicím by nám ale neměl uniknout jeho nejdůležitější závěr. Tlak Spojených států na evropské spojence nikterak nepolevil, právě naopak.Všechny díly podcastu Názory a argumenty můžete pohodlně poslouchat v mobilní aplikaci mujRozhlas pro Android a iOS nebo na webu mujRozhlas.cz.

I dagens avsnitt reser vi till 1200-talet – riddartiden, med borgar, vallgravar och torneringar! Kvällens tokroliga saga "Riddaren med den rullande pannkakan" är önskad av Alexander, 6 år från Stockholm.Riddar Rassel lagar världens största dunderpannkaka – men kastar den så högt att den landar på högkant och börjar rulla! Den rullar genom hela slottet, ut på borggården och rakt upp på en sur stenkungstaty som inte lett på fyrahundra år. Och då händer något alldeles oväntat och varmt.I studion landar tidsmaskinen vid en vit stenborg med vallgrav och ankor. Tobias har lånat en plåtrustning som är tre storlekar för stor och ligger på rygg som en omkullvält glödlampa tills Niklas drar upp honom. Plus fem fakta om riddartiden – visste du att det tog hela 14 år att bli riddare?Stötta podden och få tillgång till nya sagor! Gå med i Magiska Godnattsagor-klubben!Skicka in förslag på kommande sagor via www.magiskagodnattsagor.seSökord: magiska godnattsagor, godnattsaga, barn, läggdags, podcast för barn, barnlitteratur, ai, godnatt, riddare, riddartiden, borg, pannkaka, medeltiden, Stockholm

Před osmdesáti lety, v červenci 1946, byly veřejnosti poprvé představeny bikiny. Dvoudílné dámské plavky, na jejichž výrobu není třeba mnoho látky a které odhalují pupíkVšechny díly podcastu Planetárium můžete pohodlně poslouchat v mobilní aplikaci mujRozhlas pro Android a iOS nebo na webu mujRozhlas.cz.

Fotbalisté Francie vyřadili v prvním čtvrtfinále mistrovství světa Maroko. Zbývá tak už jen sedm týmů a šest stadionů, na kterých se bude hrát. Ty americké často stojí daleko za městem a není snadné ani levné se k nim dostat. K nejhůře dostupným patří právě stadion za Bostonem, kde Francie a Maroko v noci na pátek hrály. Kvůli dopravnímu kolapsu v okolí stadionu se tisíce fanoušků nedostaly na tribuny včas – přitom také v Americe to jde i jinak.Všechny díly podcastu Seriál Radiožurnálu můžete pohodlně poslouchat v mobilní aplikaci mujRozhlas pro Android a iOS nebo na webu mujRozhlas.cz.

Kromě hereckého talentu si do Prahy přinesla hudbu, zpěv, tradiční folklor a zkušenost s hrou na cimbál. Kvůli filmu Tanec s medvědem se ale herečka Pavla Gajdošíková musela naučit hrát na housle. „Měsíc jsem každý den trénovala dvě hodiny. Obdivuju partnera a sousedy, že to se mnou vydrželi,“ říká. Ve snímku Dřevorubec, který představila na festivalu v Karlových Varech, si zase vyzkoušela řízení traktoru i život v prostředí dřevařů.Všechny díly podcastu Host Radiožurnálu můžete pohodlně poslouchat v mobilní aplikaci mujRozhlas pro Android a iOS nebo na webu mujRozhlas.cz.

Kromě hereckého talentu si do Prahy přinesla hudbu, zpěv, tradiční folklor a zkušenost s hrou na cimbál. Kvůli filmu Tanec s medvědem se ale herečka Pavla Gajdošíková musela naučit hrát na housle. „Měsíc jsem každý den trénovala dvě hodiny. Obdivuju partnera a sousedy, že to se mnou vydrželi,“ říká. Ve snímku Dřevorubec, který představila na festivalu v Karlových Varech, si zase vyzkoušela řízení traktoru i život v prostředí dřevařů.Všechny díly podcastu Host Radiožurnálu můžete pohodlně poslouchat v mobilní aplikaci mujRozhlas pro Android a iOS nebo na webu mujRozhlas.cz.

We've been running a bit of an Agent Cloud series surveying all the top inference/compute/cloud providers, from Databricks to Daytona to Railway and, even further back, E2B, but we're excited to conclude this series returning to Modal, which has just raised a monster $355M Series C.The cloud was built for developers. But agents are now changing that.The old infra stack was designed for a human who could read docs, reason through YAML, and understand dashboards to figure out what they need when something broke. While this was painful for developers, it worked since they could fill in missing context in their heads.However, agents don't have that luxury. Now in this new era of agents, everything has to be tighter.They need a place to write code, run it, inspect the output, change the environment, debug failures, and try again. Fast iteration and feedback loops with all the necessary context are crucial for agents to operate properly. Furthermore, sandboxes are a clear representation of this shift as agents can easily spin up isolated environments. This programmatic infra even extends to research:Two years ago, we were one of the first to cover Modal with CEO Erik Bernhardsson and Alessio designed our favorite LS thumbnail of all time:At the time, Modal was just a teeny little company with a $17M Series A.Today, fresh off their $355M Series C, Modal is one of the clearest examples of the agent cloud future being built in real time: a cloud platform moving past traditional web app assumptions toward the workloads AI actually creates such as elastic inference, sandboxes, GPU burst, post-training, background agents, and infrastructure that agents themselves can operate.In this episode, Modal CTO Akshat Bubna joins swyx and Vibhu to unpack why AI applications don't fit traditional cloud assumptions, why Kubernetes was never designed for bursty compute-heavy workloads, and why Modal is now shifting from developer experience to agent experience.We go deep on Modal's AI infra stack: serverless functions, decorator-based infrastructure, elastic inference for custom models, GPU snapshotting, DeFlash, speculative decoding, Auto Endpoints, sandboxes, persistent storage, networked containers, private IPv6, RDMA, multi-node training, and Modal's capacity pool across 17 cloud providers. Akshat also explains why RL rollouts can require 100,000 sandboxes, why production agents need hard guardrails, why observability may matter more than reading code, and why AI has made infrastructure exciting again.We discuss:* Why Kubernetes wasn't built for bursty AI workloads* How Modal started as a better runtime before becoming an AI cloud* Why Modal added GPUs before ChatGPT* The shift from developer experience to agent experience* Why observability matters when agents are writing the code* Elastic inference for custom models across audio, video, robotics, and comp bio* GPU snapshotting, cold starts, and why inference workloads are so bursty* Why RL rollouts can require 100,000 sandboxes* DeFlash, speculative decoding, and frontier-level inference performance* Auto Endpoints and making optimized inference easier to deploy* What Modal adds beyond vLLM, SGLang, and raw GPU rental* Modal's 17-cloud capacity pool and supercloud strategy* Networked sandboxes, sidecars, private IPv6, and RDMA* Serverless multi-node training for post-training and research workloads* Auto-research, model-guided sweeps, and agents launching GPU experiments* Compute strategy, capacity planning, and batch tiers* Why production agents need specialized sandboxes and hard guardrails* Modal's take on managed agents, CI, Gitpod/Ona, Python, TypeScript, and Modal BenchAkshat Bubna* LinkedIn: https://www.linkedin.com/in/akshat-bubna-188885103* X: https://x.com/akshat_bModal* Website: https://modal.comTimestamps00:00:00 Introduction00:00:39 Modal's origin and why Kubernetes wasn't enough00:04:32 Developer Experience → Agent Experience00:06:21 Modal's AI cloud primitives00:09:14 Sandboxes, agent loops, and proto-Cognition00:12:12 Elastic inference, GPU snapshotting, and 100,000 sandboxes00:15:24 DeFlash, speculative decoding, and Auto Endpoints00:19:59 Production-grade inference beyond raw GPUs00:22:00 Background agents, Ramp Inspect, and the agent lifecycle00:24:08 Modal's 17-cloud supercloud strategy00:26:40 Networked sandboxes, private IPv6, and RDMA00:32:48 Multi-node training, post-training, and auto research00:37:36 Compute strategy, capacity planning, and batch tiers00:40:55 Open models, real-time AI, and production agent infra00:43:06 Hard guardrails, managed agents, and specialized sandboxes00:46:06 Why AI made infrastructure exciting again00:48:30 Model APIs, differentiated products, and agentic video00:51:50 CI, coding-agent infra, SDKs, and Modal Bench00:57:28 Closing ThoughtsTranscriptIntroduction: Modal, Series C, and the Art PartySwyx [00:00:00]: We're here with Akshat, CTO of Modal, together with Vibhu. Congrats on your Series C.Akshat [00:00:10]: Thank you.Swyx [00:00:11]: Your party yesterday was amazing.Akshat [00:00:15]: Yeah.Swyx [00:00:15]: From all the photos and all the swag.Akshat [00:00:17]: We had a bunch of art installations, which was fun, seeing, like, our products on pedestals next to, like, Rodin.Swyx [00:00:25]: Very nice. Very nice. When you started, it was not the GPU inference company. Maybe it was in your mind. Take us back to the origin story.Modal's Origin: A New Runtime Beyond KubernetesAkshat [00:00:39]: I first met Eric, who's the CEO, through an investor. Back then Eric was already thinking about building, a new runtime, and he got there thinking through why are workflow orchestration products so hard to use. It's because you have to run them on Kubernetes. Kubernetes is hard to manage. It's not built for burstiness and, custom images,Swyx [00:01:03]: YeahAkshat [00:01:03]: It has a terrible developer experience.Swyx [00:01:05]: And I'll, I'll interjectAkshat [00:01:06]: YeahSwyx [00:01:07]: For listeners, who are new, we interviewed Eric two years ago, and there's a bit more of the story there from Spotify and all those things.Swyx [00:01:14]: And I came across Eric through Data Council because he did that talk on the serverless container stack that you guys did, which was like, that was my first like, “Okay, I need to take Modal very seriously” moment.Akshat [00:01:26]: Yeah.Swyx [00:01:26]: But it was still very unclear, like, do I need all this for just my data pipelines?Akshat [00:01:33]: Yeah. initially what we were thinking about was if we build a better runtime, it's a very useful primitive in itself. It's There's a lot of things that, get solved by serverless functions, like you can do, ETL stuff, you can do job queues, you can do all this, like, bursty processing, which it turns out every company had needs for. but then we also were thinking about this as like, this is a primitive that we can build a whole collection of products on, which are very verticalized. So perhaps data engineering would've been the first one, but we were thinking about inference. Back then it was more classical inference, like computer vision stuff and running XGBoosts and whatnot. But we added GPUs to the product a year before ChatGPT came out.From Serverless Containers to GPU WorkloadsSwyx [00:02:19]: Nice.Akshat [00:02:19]: We just didn't think it would be that big of a deal.Swyx [00:02:22]: Yeah, just like add A100.Vibhu [00:02:23]: Was there any, like, early key problem that really sparked off why you built it?Akshat [00:02:28]: Yeah. Primarily it's just, none of the tooling that was out there was built for, one, a really great developer experience, and also there's a general trend of, a lot of the workloads that we were seeing were very. I wish there was a better word for it, but compute-heavy. Like, they need, one, like, need a lot more resources, so you need to burst up and down a lot, versus like Kubernetes designed for, like, slow scaling and, more for, like, web server use cases. And also there's just a lot more specialization in, like, what kinds of environments these workloads run in. Like, we had sometimes they need accelerators, sometimes they need different kinds of images, and this is just like a consistent thing that we saw across a lot of companies. That would be the next step.Software-Defined Infrastructure and Decorator-Based DXSwyx [00:03:13]: Yeah. Yeah. Be nice. I don't know how much this factored into the early story, but I wrote a post when I was at Temporal about infrastructure, software-defined infrastructure or something like that.Akshat [00:03:22]: Yeah, the self-provisioningSwyx [00:03:23]: Self-provisioning.Akshat [00:03:24]: Yeah.Swyx [00:03:24]: Yeah. I can't even remember my own post.Swyx [00:03:26]: And then you put me on the landing page.Akshat [00:03:28]: Yeah. We really like, the term and so we stole it.Swyx [00:03:32]: Because you had the insight that everything can just be in decorators co-located with the code, right?Akshat [00:03:37]: Yeah.Swyx [00:03:37]: Was that a big part of the originalAkshat [00:03:39]: YesSwyx [00:03:39]: Story or it was just like a DX layer?Akshat [00:03:41]: That was, really important because we really didn't want people to spend, so much time, writing YAML, and it seemed like you could really condense the surface area of what you're doing, put it in code so you can operate on it just like you operate on other code, and like build stuff that's more expressive and dynamic. and so yeah, that was always a very important part.Swyx [00:04:04]: Then the pushback is this is a DSL.Akshat [00:04:07]: Yeah.Swyx [00:04:07]: It's you're closed source. I am locked into Modal.Akshat [00:04:11]: Yeah. We never really got pushback for that because the nice thing about Modal is you can bring whatever code you have, and sure, the DSL is at the configuration layer for, what hardware you're using, how you're scaling things up, but you still own the code.Akshat [00:04:27]: And that's, that's been an important, part of our story, even as we do inference now.Swyx [00:04:32]: Yeah.Vibhu [00:04:32]: How much of do you think still stays the same today? Like if you were to build something today, DevX very important, but I feel like, a lot of this has been changed with just hook it up to an agent, have Claude Code, have Codex implement a tool. there's very agent native primitives that are different than if I'm doing this myself, right?Developer Experience → Agent ExperienceAkshat [00:04:54]: We've changed our SDK team to think about agent experience instead of, developer experience and we think that the same benefits that apply for DX also apply for AX, which is why would you have an agent read through hundreds of Kubernetes files and like write YAML that's not even typed when it can make a couple of changes in a decorator and it gets this self-provisioning runtime of, being able to see its changes live in action? yeah, it just seems from the customers we talk to, they find Modal is much faster for agents to use versus operating on a different substrate.Swyx [00:05:34]: Yeah, because like you, again, you co-locate the infrastructure requirements to the code that runs it.Akshat [00:05:38]: Yeah.Swyx [00:05:38]: Well, the negative thesis now is that nobody's looking at their code anymore, so there's no point.Akshat [00:05:44]: Yeah, people aren't looking at code. one thing we still see is really important is observability.Swyx [00:05:51]: Yeah.Akshat [00:05:51]: Like how good is your dashboard? And of course, like we have, we push a lot of it to the CLI so the agents can do their own investigation, but you still need humans to go interpret what's going on and, make judgment calls and whatnot. and that's I feel like, Maybe more important now than looking at the code itself.Swyx [00:06:11]: Yes, because like, you can try to treat the code as a black box and then use, see the observable action that comes out of it, and then just prompt a change.What Modal Is For: AI Cloud PrimitivesAkshat [00:06:21]: Yeah.Swyx [00:06:22]: So I think it takes a bit of restraint to not specialize, to say, “I want to ship a new primitive,” and then just be general purpose.Swyx [00:06:31]: People ask you, “What are you for?” You're like, “ I don't know. We can do this, we can do that.”Vibhu [00:06:36]: Well, I'd be curious to see, like, okay, if we were to ask you, like, what is Modal for even at a high level? There's a lot you guys do, sandboxes, GPUs, everything. How do you answer?Akshat [00:06:46]: Modal is a cloud platform that's built for, where we've built the primitives from scratch for AI applications. and right now it covers, inference, training, batch processing, and sandbox workloads.Akshat [00:07:00]: But we're building a lot moreSwyx [00:07:02]: I noticed you didn't say web server, so there is still a role for, like, the always-on large-scale Kubernetes type things.Akshat [00:07:09]: Yeah, absolutely. We're, we're not trying to compete with the renders of the world, because yeah, we think the differentiator for us is the, are the workloads that need specialized compute, need to scale up and down a lot. yeah, they're, they're, they're just shaped differently.Working Alongside Frontier StartupsVibhu [00:07:26]: I think you're building a lot of it alongside the startups, right? They're innovating quite a bit, even in your, like, latest blog post. Like, even in the series C, the customers that you mention here, the cognitions, technical ones, ramps and whatnot, they're, they're innovating with you, right? And that's not something AWS is doing directly with.Akshat [00:07:45]: Yeah, absolutely. I think, this is again classic. We're a small team. We can move really fast. our engineers are working with our customers and figuring it out. Yeah.Swyx [00:07:54]: So my first week at Cognition, I walked in, there was someone wearing a Modal shirt. I was like, “What are you doing here?” They're like, “Yeah, I just. I am embedded inside of Cog.”Akshat [00:08:05]: Yeah, I think that was Peyton. We sent him overSwyx [00:08:07]: Yeah.Akshat [00:08:07]: Because, the latency of communication was too high otherwise.Swyx [00:08:12]: Yeah, distributed node, you have to - you have to place one and collocate.Vibhu [00:08:16]: Yeah.Swyx [00:08:16]: So I had a, I had direct personal experience, right? So I worked on smol developer three years ago. it was inspired by Claude 1. I think you onboarded me at some point, like, just before, and I was like, “Oh, like, I need some bursty compute. Like, I was just gonna try using Modal.” And it was a, it was a pretty pleasant experience. apparently, I showed up in the board meeting, like the analytics.smol developer, Sandboxes, and Proto-CognitionAkshat [00:08:39]: Yeah, you blew up on Hacker News and,Swyx [00:08:41]: YeahAkshat [00:08:41]: We got a big traffic spike. I. I think the way you used smol developer was Modal functions for running stuff, which was. Like, the, that was a good use case. but then, yeah.Swyx [00:08:53]: Yeah. That - So to me, that was proto-cognition.Akshat [00:08:55]: Right.Swyx [00:08:56]: If only I had, like, stuck to it.Swyx [00:08:58]: Like, that was like, if - did you say draw the tech treeAkshat [00:09:00]: AbsolutelySwyx [00:09:00]: You're just like, “Yeah, like, probably this will happen.”Akshat [00:09:02]: Yeah. Like, he was so close. You were just rebuilding upon usSwyx [00:09:04]: I just didn't realize.Akshat [00:09:05]: But the funny story there is at the same time, we were talking to a bunch of customers who needed something like sandboxing.Swyx [00:09:14]: Yeah.Akshat [00:09:14]: This is like twenty-three.Swyx [00:09:15]: Yeah.Akshat [00:09:16]: So we builtSwyx [00:09:17]: You introduced a new API right after that.Akshat [00:09:18]: Yeah.Swyx [00:09:19]: Yes.Akshat [00:09:19]: Like, we built sandboxes in May of twenty-three before anyone was even knew this was gonna be a thing. And the first example we published was, we took smol developerSwyx [00:09:28]: Smol developerAkshat [00:09:28]: And put it in a loop, so the agent can iterate on itself.Swyx [00:09:33]: Loops are hot these days.Vibhu [00:09:34]: It's the looper.Akshat [00:09:34]: Yeah.Vibhu [00:09:35]: Loops in. When was this, twenty-three?Akshat [00:09:38]: Yeah.Vibhu [00:09:39]: A small check.Akshat [00:09:39]: Yeah.Swyx [00:09:39]: It's like twenty-three. so the. the, those for listeners, like, the problem was the models are not built for any of this, right?Swyx [00:09:46]: Like, you're just trying to like. They're not post-training to understand, like, looping and, like, self-correction and tool calling was there, but, like, also not that great.Akshat [00:09:55]: Yeah.Akshat [00:09:55]: I don't remember if you used tool calling in this one, but yeah, the models would just diverge after like ten iterations and not produce anything meaningful.Swyx [00:10:03]: Yeah. But like, then. So okay, like now talking to myself three years ago, the answerVibhu [00:10:08]: Of course they will get betterSwyx [00:10:09]: Collect all the failures, build benchmark, and then collect all the, examples, build the RL environmentAkshat [00:10:15]: RightSwyx [00:10:15]: Sell it for like ten billion dollars to Meta.Swyx [00:10:17]: And then also train a model and then sell that for sixty billion dollars to Elon. And this isAkshat [00:10:23]: Yeah, of courseSwyx [00:10:23]: The funny machine. Like, it's like, it's about the hardware.Akshat [00:10:28]: It's hard to have that inherent conviction that the stuff will get that much better.Swyx [00:10:33]: In retrospect, it's so f*****g obvious.Akshat [00:10:36]: Fair enough.Swyx [00:10:37]: Like, what else were we doing back then? I don't know. anyway. Yeah. So this. That was the start of your sandboxing journey, right? I feel like it didn't blow up until, like, last year.Akshat [00:10:49]: Yeah.Swyx [00:10:50]: So there was like a couple years of quietness.Akshat [00:10:52]: Exactly, yeah. We wereVibhu [00:10:53]: I think very underrated product value. Like, my experience with Modal, Charles, before he had joined Modal, met this guy at a hackathon, and he really insisted we wanted to run some small model, not hosted anywhere, and he's like, “ there's this cool company, Modal. They'll like spin up a GPU sandbox, we can throw it on there. They'll take a Hugging Face link.” And like there's so much value just right there, right? Like instant hosting, spin it up, spin it down. It'll stay cold, but we run the demo a few days later, it'll come back up and like all this stuff in retrospect, like it's still what we needed like today.Akshat [00:11:27]: Yeah, it's still needed today. workload shapes have changed a lot as, we run stuff for people with really massive production scale and, there it's it's not about scaling from zero to one, but it's how do we scale really elastically, from like thousand to fifteen hundred GPUs very quickly in a given region. It's the same shape problem.Elastic Inference, GPU Autoscaling, and Custom ModelsVibhu [00:11:50]: Okay. So you look at, say, Cursor Composer, right?Akshat [00:11:53]: Yeah.Vibhu [00:11:53]: They had a. “We'll do RL on a model every couple hours.” you guys have a whole version of RL inference gym and whatnot.Vibhu [00:12:01]: When you look at workloads like that, you're doing train runs where you need to scale up, scale down every hour thousands of GPUs, right? That's the example for we do need it, right?Akshat [00:12:12]: Yeah. Well, so I'll, I'll take a step back and, maybe talk about like how people use Modal today. because our biggest use case is, elastic inference. And the thing we first found product market fit, with was inference for custom models. So we stayed away from the LLM space, and we were serving companies like Suno for audio, Runway for video, robotics, comp bio companies that train their own model elsewhere. But Modal is the best black box that for deployment, scaling to however many GPUs you need as your traffic pattern changes. And we saw all of them like have a very unpredict- predict- predictable, traffic pattern. it's like diurnal. It's Some days, like the company will do a launch and, they'll need like, way more. And it's not just one model that they deploy. They-- all these companies deploy, lots of different models in different regions, and so the autoscaling problem becomes even harder because then you have to scale within a certain region, and those cycles are offset. So different times you scale up in different regions.Akshat [00:13:20]: So that's like our sortVibhu [00:13:22]: And thatAkshat [00:13:22]: YeahVibhu [00:13:22]: That in and of itself is a huge category. There's a bunch of inference providers which, provide this fireworks, does this as a service together, whatnot, Base10. that's carved into its own niche for language models, at least right now.Akshat [00:13:36]: Yeah. the thing that we have specialized in is the autoscaling aspect.Vibhu [00:13:41]: Yeah.Akshat [00:13:41]: Because we found that it's not universally true that everyone else can autoscale, and we've gone deeper into it on the tech side by, we've incorporated GPU snapshotting into the product so we can take the GPU state, like your torch.compile model, snapshot it, and the next cold start is way faster. And so going back to your question, it's That's why you need a lot of burstiness for inference. But then people also do a lot of demand training, like for RL stuff, your rollouts are bursty, as you said. People also do a lot of batch jobs. So we'll see, a lot of companies, before they have a training run, they'll need thousands of GPUs to run encoding or something like that. And I think those things are much more bursty than. I agree that agents are not that bursty. sandboxes are, except when you're doing RL. RL is justRL, Batch Jobs, and 100,000 SandboxesVibhu [00:14:28]: Or commerceAkshat [00:14:28]: Insanely bursty.Vibhu [00:14:29]: Yeah.Akshat [00:14:30]: Yeah. Like when you're doing, rollouts, you sometimes need a hundred thousand sandboxes in your sandboxes.Vibhu [00:14:37]: Yeah. I'm curious if you've seen early sparks of continual learning. There are some people, like our friends, ngram, recently announced thisAkshat [00:14:45]: YeahVibhu [00:14:45]: They're, they're trying to do training. That also seems like a different workload, right? If you're doing training twenty-four/seven per se, there's a very weird dynamic of how you're using GPUs between people and whatnot, but seems like something you guys would work for.Akshat [00:15:00]: As you said, we're, we're fortunate to work with a number of, customers at the frontier and grab some of our customers. and they are taking the primitives we have, and trying to use them in very interesting ways, like continual learning. It's possible as the stuff gets better, some of that will be part of, our offering as well if, more people need it. but we're, we're just waiting to seeVibhu [00:15:23]: YeahAkshat [00:15:23]: How it shakes out.Vibhu [00:15:24]: Is there a primitive that you added after sandboxing that was the next step in the story?LLM Inference, DeFlash, and Speculative DecodingAkshat [00:15:32]: I guess we've been going much deeper into LLM inferenceVibhu [00:15:35]: YeahAkshat [00:15:35]: Because we realized that some of the advantages we have with like autoscaling, again, especially in different regions and whatnot, are, not present elsewhere. and the place where we had a gap was we weren't, working on the model layer itself. Like we were a black box. And, we realized that, we can get to frontier-level model performance, with, by having great people who work on this. And, we've been open sourcing a lot of our work, in terms of, Recently, we, shared our work on DeFlash, which is a block-based, speculator, and we've open sourced, all of it. So, you can - By using open source DeFlash, you can get the same performance as you would with one of the proprietary providers. And the next thing we're thinking about hereVibhu [00:16:23]: I thought this wasAkshat [00:16:24]: YeahVibhu [00:16:24]: An interesting blog post as well, right? Like, I think in here you make a claim that. Not a claim, just that how effective speculative deco-decoding really just get to.Akshat [00:16:33]: Yeah.Vibhu [00:16:33]: Anything you wanna point out from this around, what people should know?Akshat [00:16:39]: Yeah, absolutely. the high-level summary is, it would help to describe what speculative decoding is.Vibhu [00:16:44]: Yes.Akshat [00:16:44]: I will, yes.Vibhu [00:16:45]: I think, likeAkshat [00:16:46]: YeahVibhu [00:16:46]: So we've covered like Eagle and all thisAkshat [00:16:47]: YeahVibhu [00:16:47]: Like Hydra and all those things, but it was like two years ago.Akshat [00:16:51]: Yeah.Vibhu [00:16:51]: I think it doesn't hurt, right?Akshat [00:16:52]: Yeah. Speculative decoding is you have a smaller model, called a draft model, predict tokens ahead of the bigger model, and then you have the bigger model, verify all of this, all the tokens are predicted. And the reason it's faster is if you're predicting, one token at once, you're bound by memory bandwidth. But if you can batch the verification of, the draft model, then you're much more efficient using compute, and it's faster, and as long as your draft model is producing a lot of tokens that can get accepted, which is called the accept length, you can get a speed up that's, multiple times of, the original model speed. and well, that's what we highlight here. It's Like people talk a lot about we made these kernels faster and whatnot, but improving kernel will only give you like few percentage points of improvement, and, increasing accept length, literally is a multiplicative decreaseVibhu [00:17:47]: Like two to four X.Akshat [00:17:48]: Yeah, exactly.Vibhu [00:17:48]: Without much head-on performance.Akshat [00:17:50]: Yeah. I think it may - you are running a second model, right? So it may be something more expensive in the compute,Vibhu [00:17:57]: I meant quality performanceAkshat [00:17:58]: Probably not by muchVibhu [00:17:58]: But yeah. I thinkAkshat [00:17:59]: So there's no drop in quality performanceVibhu [00:18:01]: YeahAkshat [00:18:01]: Because you're always. You're never accepting a token that the big modelVibhu [00:18:04]: It's strictly betterAkshat [00:18:05]: YeahVibhu [00:18:05]: Or it's same.Akshat [00:18:06]: Exactly.Vibhu [00:18:07]: Right. Yeah.Akshat [00:18:08]: And so we've been working a bunch on DeFlash, which is a block-based speculator. so it's instead of predicting, one token at a time, it's predicting a block. And we've been open sourcing our work with it. The next thing for us here is for helping people train speculators and custom models. it's it's something that traditionally is very forward-deployed engineering driven, support deployed, engineer driven, like you work with customers and help them do that. And our vision for. This is why we launched Auto Endpoints, is we want to make frontier-level performance available to everyone. And so, we mentioned this in the announcement, we teased it. The next thing we're, we're launching is, as you run an auto endpoint, we shadow trafficAuto Endpoints and Frontier-Level PerformanceVibhu [00:18:54]: Do you want to explain what auto endpoints are?Akshat [00:18:57]: Yeah.Vibhu [00:18:57]: I lovely, yeah.Akshat [00:18:58]: Yeah. So, this is, I guess, going back to your Modal is you touch the code, but, sometimes people don't wanna touch the code, and they wanna get started with an endpoint that works and has all the great performance and, scalability that Modal has. So we've made that easier with, a way to create an endpoint from our UI, from the CLI, that has all of our optimizations that we talked about, like the DeFlash stuff already baked in, and there's full transparency. So we give you the code, you can go run it yourself, and if you want, you can eject out into the full Modal experience, which we see as people get sophisticated, they do wanna tweak the models, they wanna, fine-tune stuff. You can still do all of that. It's it's not a black box. And yeah, the next thing, as we teased later in the post, is how do we give you value even beyond this in terms of having your draft models evolve as your data distribution evolves, again, without having to talk to a person and, yeah.Vibhu [00:19:59]: I guess just to understand it directly, you have the GPUs, you have an endpoint that's compatible, you serve open model. If someone was to do this themselves, what's the delta that you guys provide? So you do a lot of open source great work on effective inference. how does it compare to, say, I take the same model, 5.2 FP8, take shelf inference engine, vLLM, SGLang, get compute of similar capacity, similar cost. What's the delta that plugging into something this, like this offers outside of the benefit of, scaling?Production Inference Beyond Raw GPUsAkshat [00:20:34]: It's interesting because we've taken the approach of open sourcing our contributions and upstreaming them. we work closely with the SGLang team. We want the improvements that our team, comes up with to be, there in open source for others to use, even outside of Modal. The benefit to us is we have a team that has significant expertise in terms of if you do have something that is not there, our team can help you get that performance, first. the other thing is with these endpoints, we are way more elastic, as you said, than, anyone else, and you have true scaling to zero. you have true, burstiness, and in practice, that matters a lot more to people than just finding, the GPU and, running Modal code on something.Vibhu [00:21:20]: Yeah. And I will say it's not that straightforward to just. like what I said is easier said than done, right?Akshat [00:21:26]: Yeah.Vibhu [00:21:27]: It's I think still for the average person, still hard to just gut check using different. There's, there's quite a bit of combinations you can make there. the trade-offs aren't really known at face value.Akshat [00:21:40]: Yeah. it's it's not just that. I think it's it's that running production-grade inference is a hard infer problem.Vibhu [00:21:49]: YeahAkshat [00:21:49]: Even if you subtract out the autoscalingVibhu [00:21:50]: YeahAkshat [00:21:51]: Is controlling things like tail latency and, making sure every, request is delivered at least once and whatnot.The Model and Agent LifecycleVibhu [00:22:00]: There's a lot of innovation that you can do here. I think, it's very interesting that you're starting to encroach on, like as you become a full cloud, you're starting to encroach on other people's turf.Vibhu [00:22:09]: What will you not do?Akshat [00:22:13]: Well, we wanna follow our users and, make sure they get like a platform that has everything that works well together. so right now we're focused on the model lifecycle and the agent, lifecycle. so both like going from data prep to training to inference, and then also if I want to deploy a background agent, let's say, sandbox, do persistent storage, a whole bunch of other stuff.Vibhu [00:22:38]: We talked to Cole, who did, OpenInspect. Yeah.Akshat [00:22:42]: Yeah.Vibhu [00:22:42]: And RealInspect also is on Modal.Akshat [00:22:44]: Yeah. So Ramp Inspect was a great example of a background agent that was really successful because they, were able to use some of the primitives like snapshotting and fast scaling to just have something that feels really reactive and works well.Ramp Inspect and Background AgentsVibhu [00:23:02]: Yeah. That's the new CTO of, Ramp right there.Akshat [00:23:05]: Yeah, Rahul.Vibhu [00:23:08]: It was really fun. yeah, okay, I think, all very bullish. Like, one of my reflections was also I did not originally. So when I met you guysThe Inference Inflection: CPU, GPU, and Co-LocationVibhu [00:23:19]: You weren't that much in the GPU game, and now you're all about, inference. And one of the points that I hinged on for Jensen's keynote at GTC this year was, what we're calling like the inference inflection, right? That let's say in AI workloads or machine learning workloads, it used to be like, let's call it eight to one GPU to CPU, and now it's more like one to one, which is like a interesting. Like, - because of how much agents are blocked or call out to this, to CPU heavy stuff the actual, like, limiting factor, like, swings back and forth from GPU to CPU a lot more than it used to be all GPU and then occasional CPU.Akshat [00:24:01]: Yeah.Vibhu [00:24:02]: GPU, CPU. And now it's like just constantly, and you just have to locate everything.Seventeen Clouds and the Supercloud StrategyAkshat [00:24:08]: Yeah. And that's one of the things that, again, we see as, something appealing about Modal, which is we've built this capacity pool that spans, 17 cloud providers, so we're, we're very good at Running on various kinds of cloud capacity across the worldSwyx [00:24:24]: You don't have your own data centers?Akshat [00:24:25]: We don't have our own data centers. We just run across a lot of neo cloudsSwyx [00:24:29]: Yeah. AreAkshat [00:24:30]: Metal providers.Swyx [00:24:30]: Yeah. Question mark.Swyx [00:24:31]: Yeah. You're, you're running the math, and you're like, “What's the cutover point where you're like.”Akshat [00:24:36]: Yeah, it's a good question. part of it is we see our differentiator in the software layer, and, being capital light and focusing on the software helps us move really fast. so far it's worked out well because there are so many other people building data centers that we're able to work effectively with them, and again, focus on what makes us, special.Swyx [00:24:55]: Yeah.Swyx [00:24:56]: 17 gets you into, like, the local providers sometimes. LikeAkshat [00:25:00]: The,Swyx [00:25:01]: Which was the most interesting one?Akshat [00:25:02]: There are a lot more neo clouds than you expect, and they all have various degrees of, various levels of reliability. And, that's why it's something we've invested a lot of time in, is building our own reliability layer on top. so if the GPU falls off the bus or something happens, we user workloads are not affected, and that lets us use a lot more capacity than,Swyx [00:25:30]: YeahAkshat [00:25:30]: You as a user would be able to.Swyx [00:25:32]: It's a useful thing to have because like now everyone knows, like, what layer you are and, like, you optimize for being the super cloud of all clouds.Akshat [00:25:41]: Yeah. That's, that's, that's the idea. and so I guess when you mentioned colocation, that's, that's another interesting thing where, one thing we've seen is people come to us when they want, very specifically located, CPUs or GPUs, like they wantSwyx [00:25:57]: Oh, they pin it in likeAkshat [00:25:58]: YeahSwyx [00:25:58]: EU?Akshat [00:25:59]: Exactly. Or EU, US.Swyx [00:26:01]: Right. Data resiliencyAkshat [00:26:02]: AustraliaSwyx [00:26:02]: Locality thing or performance or what?Akshat [00:26:04]: It's either data locality or latency, yeah.Swyx [00:26:07]: Yeah.Akshat [00:26:07]: Like, you want your. They're running sandboxes and model. They want them to be right next to aSwyx [00:26:10]: Yeah, it's easy thenAkshat [00:26:11]: YeahSwyx [00:26:12]: To. That is important in all those things. and so, like, you've accidentally, I don't know if it's accident, but, like, you've built the perfect primitive for agents to express themselves. And then, like, it's almost very funny how every extra development just involves more file system, just involves more CPU.Akshat [00:26:30]: Yeah.Swyx [00:26:31]: Just like the things that you already have. I don't know much about, if there's any, like, networking usages that are interesting, but you've also done some good work on networking.Networking, Sidecars, Private IPv6, and SandboxesAkshat [00:26:40]: Yeah, that's exactly right. Like, we're just taking compute storage and networking and building stuff on that layer, for, again, the stuff people need.Swyx [00:26:49]: YeahAkshat [00:26:50]: We see a few interesting networking things coming up. one is people want networked sandboxes. so we haveSwyx [00:26:57]: For like a Docker cluster type thing.Akshat [00:26:59]: Yeah.Swyx [00:26:59]: Sorry, Docker Swarm. Oh, f**k. What is it called?Akshat [00:27:02]: Compose.Swyx [00:27:03]: Compose type thing.Akshat [00:27:04]: Yeah. So if you want Docker Compose, our sandboxes now support, this thing called sidecars. So you can. A sandbox is a pod of containers, and you can run multiple containers in, a sandbox. also useful because, going back to networking, people want a lot of control over, outbound networking from a sandbox.Swyx [00:27:23]: Yeah.Akshat [00:27:23]: Like, they might wanna run a middle proxy for, like, maybe logging stuff for RL or, controlling how egress can happen to a domain, injecting credentials. and yeah. So we've, we've had to build a lot of that stuff ourselves.Swyx [00:27:38]: Yeah.Akshat [00:27:39]: But then also sometimes people want, sandboxes spanning multiple nodes to talk to each other, which is an emerging thing we're seeing. We have support for that for a different reason, and yeah, we'll see if that becomes stable.Swyx [00:27:52]: Like, just an open socket. It's a. This is directly like mTLS.Akshat [00:27:56]: We do support that, which is you can, expose a tunnel inside a sandbox.Swyx [00:28:01]: Yeah.Akshat [00:28:01]: And then you can either expose it to public internet or it can be, you can add like a HTTP, auth layer above it. But we have this thing called I6PN, which we haven't talked about, which is this, like, overlay network using IPv6 addresses. so if Modal containers, within the same workspace, when this is enabled, can address each other using this private IPv6 address, and no one else can.Akshat [00:28:28]: So it's like private networking, for containers. We built it because we needed it as a primitive for our distributed training product. so we have this other feature, which is you can add a decorator to a function, and you get a cluster of GPUs. and they have RDMA networking. so you can run a distributed training job, that's truly serverless. and we did the overlay network for that. But then we've seen that people are using it for other reasons, and, I'm intrigued to yeah, what would people do with it.Swyx [00:28:59]: Build primitives and let people figure it out, right?Akshat [00:29:01]: Yeah, exactly.Swyx [00:29:02]: You put out a pretty interestingAkshat [00:29:03]: They're like, they read the docs webpage. Let me use thatSwyx [00:29:06]: YeahAkshat [00:29:06]: Something they never intended to work. This is literally not even in our docs page. People somehow found it, and they're using it.RDMA, Memory Movement, and Distributed TrainingSwyx [00:29:12]: Huh.Swyx [00:29:14]: The way you portrayed it with, like, RDMA versus TCP, like, very well laid out, but just the transfer speed change at scale for RL, like yeah, you have it, you have it built in. I'm sure someone found it. It's found it to be a lot more efficient before you made a thing out of it, right?Akshat [00:29:32]: Yeah. And not to split hairs, I guess the overlay network is the TCP overlay network.Akshat [00:29:39]: The reason we have that is you need that to do the key exchange for RDMA before you set up the RDMA network on top of that. but then people found the TCP part.Swyx [00:29:48]: Can I tell you, this is like a big aha moment for me becauseAkshat [00:29:51]: YeahSwyx [00:29:51]: So I review 2,200 submissions for the World's Fair.Akshat [00:29:56]: Yeah.Swyx [00:29:57]: And then I got this from John OsterhoutAkshat [00:29:58]: HuhSwyx [00:29:59]: Who I don't know if. Do John Osterhout by name?Akshat [00:30:01]: The name sounds familiar.Swyx [00:30:02]: He published a. He's a well-known professor, published a lot of interesting software design books, and this is the talk he chose to submit, is on RDMA at Inference. And I'm like, you wouldn't think that this guy, who is like operating systems guy, would care about RDMA.Akshat [00:30:20]: I, it makes sense to me because I,Swyx [00:30:24]: This is the cloud, right? YeahAkshat [00:30:25]: Like, the way you move around your KV cache and how efficiently you can do it, how efficiently you move, your weights from your training GPUs to your inference GPUs in RL is there's a lot of degrees of freedom, and it is a systems problemSwyx [00:30:41]: YeahAkshat [00:30:41]: Moving memory aroundSwyx [00:30:42]: YeahAkshat [00:30:43]: Scheduling.Swyx [00:30:44]: This shows you how primitive my understanding of networking stuff is.Swyx [00:30:46]: Is this like the domain of WireGuard as well?Akshat [00:30:50]: Not quite.Swyx [00:30:51]: It's adjacent?Swyx [00:30:53]: Explain everything.Akshat [00:30:54]: Sure.Swyx [00:30:56]: How do we move memory around GPUs?Akshat [00:30:58]: Well, so sorry. Yeah, that is memory. Sorry, I was talking more, and maybe I was talking like five minutes back, about the private IPv6, addressing that you've set up.Swyx [00:31:09]: Yeah.Akshat [00:31:09]: Is it like it's a VPN?Swyx [00:31:10]: Yeah, it is like a VPN, and yeah, WireGuard is, yeah, you're right. It is,Akshat [00:31:16]: Right. Yeah, you already moved on to new topicsSwyx [00:31:17]: A similarAkshat [00:31:18]: OkaySwyx [00:31:19]: In the same space, WireGuard is, encrypted and this is,Akshat [00:31:23]: And you don't need encryption.Swyx [00:31:23]: Yeah.Akshat [00:31:24]: Yeah.Swyx [00:31:24]: This is not encrypted. that's the main difference. This is TCP and we have eBPF programs that will reject or allow the TCP connection based on whether you're allowed to do it.Akshat [00:31:35]: Used to involve a full sidecar, but now you have eBPF in the Linux kernel.Swyx [00:31:39]: Yeah.Akshat [00:31:40]: Yeah. I don't know if this is a natural follow-on to the topic of like my skepticism on distributed training is that while, like, people spend a lot of money on, like, cables to hook up GPUs, and even that is not, like, fast enough, and that's the bottleneck, is your networking fast enough?Swyx [00:31:59]: Yeah. So I guess you're talking about fully distributed training like, Dialog or something which is like cross data centerAkshat [00:32:06]: That would be, yes.Swyx [00:32:07]: That's the extreme.Akshat [00:32:08]: Yeah.Swyx [00:32:08]: You're in the middle, and then other people would have like the Mellanox cables up in, like, their actual data center.Akshat [00:32:14]: When you run multi-node training on Modal, RDMA, I think Mellanox, is, or InfiniBand is like a, is all seen as RDMA. but it's a way to bypass the TCP networking stack and, transfer, stuff much faster, between one node, to the other. And we have I think like 3 terabit per second, internal networkingSwyx [00:32:40]: OkayAkshat [00:32:40]: Which is the standard that's needed.Swyx [00:32:42]: Okay. So I misunderstood whatAkshat [00:32:43]: 50Swyx [00:32:43]: What part of the stack you wereAkshat [00:32:44]: 50 gigs overSwyx [00:32:45]: YeahAkshat [00:32:45]: If you wentSwyx [00:32:45]: YeahAkshat [00:32:46]: RDMA.Swyx [00:32:46]: Okay.Swyx [00:32:48]: Yeah. I, very impressive work.Multi-Node Training, Post-Training, and Auto ResearchSwyx [00:32:52]: So effectively you're extending like the model philosophy to the training cluster, like, yeah.Akshat [00:32:59]: Yeah. And we're, we're not going for like large scale training runs. the thing that we've built multi-node training for is, we see a lot of, smaller scale post-training. like, people are post-training like medium sized fund models, so they can, get higher quality on inference. this is a perfect fit, for something like that.Swyx [00:33:21]: Yeah. That is my impression of how a lot of these labs explore branches in post-training and then eventually merge whatever they find in.Akshat [00:33:31]: Yeah. The other use case we've seen for multi-node training is even if you have a big cluster, your researchers are still doing small runsSwyx [00:33:38]: YesAkshat [00:33:39]: Having elasticity thereSwyx [00:33:40]: Right, sureAkshat [00:33:40]: Matters a lot more.Swyx [00:33:41]: Yeah. the, like, this is like the current limiting factor for auto research, which is like you need to give your model some GPUs in order for it to completely run.Akshat [00:33:51]: We have a blog post on auto resource and model is,Swyx [00:33:55]: YeahAkshat [00:33:56]: Yeah, like, turns out to be pretty good substrate for that.Swyx [00:33:59]: So my impression is auto research means many things, likeAkshat [00:34:01]: YeahSwyx [00:34:01]: Anything that Andrej coins. Right now it's still science fair, right? Like not like, I don't know how many people are doing this.Akshat [00:34:08]: We're having a golf.Swyx [00:34:08]: Yeah.Akshat [00:34:09]: I thought the same thing.Swyx [00:34:11]: Yeah, you would know.Akshat [00:34:12]: We, like, our internal both training and inference teams use this the general shape of this quite a bit. like we have this one internal repo called auto inference, which essentially we've automated our own forward-deployed engineering efforts using, this harness, which is, the agent will just spin up a sweep of different things. It'll even run like, NVIDIA inside profiler and it'll like tweak configs and it'll arrive the right thing. it'll change your GPUs both from H200 to B200, and works really well.Swyx [00:34:47]: Nice.Akshat [00:34:47]: So yeah.Swyx [00:34:48]: By the way, I enjoy that your forward-deployed engineering is so technical that you have to do these things.Swyx [00:34:52]: It's very different from forward-deployed engineering from other people.Akshat [00:34:54]: Yeah. For our forward-deployed engineering team is, essentially they're like applied inference researchers or applied training researchers.Swyx [00:35:02]: Someone told me like they have to be able to build, but they also have to be able to sell. do they have to sell or are they like they're good, they're just like post-sale type of thing?Akshat [00:35:09]: It does, being able to talk to a customer and engage effectively with themSwyx [00:35:13]: YeahAkshat [00:35:13]: Matters a lot.Swyx [00:35:14]: They want the same thing.Akshat [00:35:15]: Yeah.Swyx [00:35:15]: ?Akshat [00:35:15]: But it's it's not really a sales, thing. We pair them with-- We have solution architects as well that are more on the sales side.Swyx [00:35:23]: Okay. Let's spend a bit more time on auto research. This is a big focus for for this year. Where does this go? like, have people explored enough? Like, there's all these beautiful charts of like improve and then level off a bit and then you find the next thing. Is this one abstraction up from normal training? Is that how we think about it, or do you think about it differently? Like model level training versus high, like driven hyperparameter search.Auto Inference and Modal BenchAkshat [00:35:51]: Yeah, like,Swyx [00:35:51]: Someone, some people call it like neural architecture search or whatever, right? Like.Akshat [00:35:54]: Yeah, - So the stuff I've seen people do with it is nowhere on the architecture level. It's pretty much tweaking parameters, but it's it's a hyperparameter sweep that's guided by some model intuition, so it's like much more efficient than, whatever other, sweep you would have.Swyx [00:36:12]: Yeah, it's just, it's just a question of where you want to spend your compute?Akshat [00:36:16]: Right.Swyx [00:36:16]: ‘Cause yeah, you can just throw infinite amounts of money on this and somehow you'll bang out Shakespeare?Akshat [00:36:22]: Yeah, infinite monkey.Swyx [00:36:24]: Yeah, so like the very good for model. and I think it's also very important that agents can spin up other agents, can spin up their infrastructure. Like very good for you. how good is our LLMs at generating model code? Like the benefit of existing LLMs is that you are in the data.Akshat [00:36:42]: Yeah. They're, they're surprisingly good. I think like pre Cloud 4 they were not, and then now they're able to shot, stuff out of the box. But we're playing around with releasing like a Modal Bench for like the harderSwyx [00:36:55]: YeahAkshat [00:36:55]: Things, that the LLMs cannot do yet and maybeSwyx [00:36:59]: What's an example of that?Akshat [00:37:01]: I think the things that- Sometimes agents struggle with, without right guidance and a skill is, how to, use the rest of our observability. Like how to. Something is failing, like how do you look at the logs and then update the right thing? It's reasoning about that. But they're able to shot, likeSwyx [00:37:23]: Yeah. You can just add a skill to it?Compute Strategy and Capacity PlanningAkshat [00:37:26]: Yeah. So we have a Modal skill now that. Which is why we built this Modal Bench. It's to find things like that, so we can address them in our tool.Swyx [00:37:35]: Tune a skill. Yeah.Akshat [00:37:36]: Yeah.Swyx [00:37:36]: No. it's it's good. are you facing any shortages? like we talk a lot about GPU shortages, but also CPU, also memory.Swyx [00:37:44]: Yeah.Akshat [00:37:45]: We have had a lot of growth, which means that, there's - we've had to be much better aboutSwyx [00:37:53]: PlanningAkshat [00:37:54]: Proactive capacity planning.Swyx [00:37:55]: Yeah.Akshat [00:37:55]: So we have,Swyx [00:37:57]: Which by the way, like it's like a MBA's like dreamAkshat [00:38:00]: YesSwyx [00:38:00]: Is like just planning this stuff. I think last time you and I talked about something maybe about this.Akshat [00:38:03]: Yeah. we have a really competent team of people that we call, The role is called compute strategy. so yeah, if anyone listening here or wants to work on thatSwyx [00:38:13]: Compute strategy?Akshat [00:38:13]: Yeah.Swyx [00:38:14]: I think,Akshat [00:38:14]: I feel like,Swyx [00:38:15]: I think the normies call it FP&A or something.Akshat [00:38:18]: Well, it's more It's it's not FP&A. It's it's There's a lot of interesting financial questions of like what is the blend between one year and three-year reservations? how do we forecast our own capacity? how do we. especially since our capacity is very fungible across different GPU types and different regions, like you have to model a lot of it. and you also have to have an opinion on how the supply chain is gonna evolve, and then you have to like, take bets,Swyx [00:38:49]: YeahAkshat [00:38:49]: Based on that.Swyx [00:38:50]: Tokenomics.Akshat [00:38:50]: Yeah.Swyx [00:38:51]: This is like probably a not a real point, but, I was trying to think about like what other industries. I was trying to think about like, we cannot be first to like these kinds of problems.Akshat [00:38:59]: Yeah.Swyx [00:39:00]: And what other industries have had this? And I was like, airlines with fuel and like they have to hedge their fuel and like, I think for a long time Southwest because they made like a hero fuel bet, they like were like super low cost becauseAkshat [00:39:12]: OhSwyx [00:39:12]: Compared to everyone else.Akshat [00:39:14]: Yeah. I hadn't thought about that.Vibhu [00:39:16]: We're at a fun time too?Akshat [00:39:18]: Yeah. It's. A lot of the compute business in general, for us is also about being very good about capacity management. That is how you have great unit, economics. but also over time it's how you can unlock more value for customers. Like, one of the things we're building now is like a way for customers to get, If they don't care about latency, like get much cheaper pricing and they'll get results back in like next 24 hours or something, like a batch tier essentially.Batch Tiers and Latency-Insensitive WorkloadsSwyx [00:39:47]: Yeah.Akshat [00:39:47]: And those are levers we have because we control the whole stack and scheduling and whatnot to give people a sufficientSwyx [00:39:53]: Yeah. I feel like they're not as popular. Like those, like the Frontier Labs have all those APIs. They're not as popular as they should be.Akshat [00:40:00]: The demand that we see for something like that is not for LLMs. although sometimes people wanna run evals andSwyx [00:40:08]: OkayAkshat [00:40:08]: Synthetic data prep and there it makes sense.Swyx [00:40:10]: Okay.Akshat [00:40:11]: But it's from a lot of LLM companies, like people who are doing computational bio, like they have to run really big batch jobs and they don't care about when they get it back.Swyx [00:40:22]: Yeah. And like they have a reasonable. It's it's also like a cousin to the stopping problem of like, will this finish in time?Akshat [00:40:30]: Yeah. You can bound it.Swyx [00:40:33]: Yeah.Akshat [00:40:33]: Like you can give peopleSwyx [00:40:34]: YeahAkshat [00:40:34]: SLAs on it.Swyx [00:40:35]: Yeah. I think what's, what's interesting is like the next phase of model.Swyx [00:40:38]: Like what, do people expect from you, now that you're established and you're like well-known compute player among all these leading companies. You had an inference launch week, and we talked a little bit about the launches. like what else? Like what else should people know?What Modal Builds NextAkshat [00:40:55]: We are building primitives that make our users' lives much easier. So, I think for example, with LLM inference, thousands more companies are gonna post-train their own models and, deploy open source models for inference. so we're thinking a lot about what is the best product shape for that. And, that involves everything from our training gym to, then, endpoints that get frontier-level performance. again, but I haven't talked to anyone. It looks somewhat different on other verticals. Like, we're also seeing a lot of real-time, audio-video stuff in there, which is why like, we're working on things like regional routing, with fallbacks. So you can get GPUs that are as close to users as possible. so you get like low latency for video streaming and whatnot. And then on the agent side, it's,Akshat [00:41:52]: We're still working very closely with our customers because stuff is changing so fast in terms of what they need. And, I think beyond sandboxes and persistent file systems, there's a lot of other things people will need from this agent stack as they build production agents. So yeah, we're thinking about those other things that fit in there.Swyx [00:42:13]: I want to ask what the other things are.Akshat [00:42:15]: Yeah. I probably should share right now.Swyx [00:42:17]: I think-- I think, okay, so, I do think a lot about the principal components of cloud, and you do talk about compute storage networking.Akshat [00:42:25]: Yeah.Swyx [00:42:25]: Because so far for me, it's fine. so far for the. the first couple generations of cloud, it's fine. What's different, qualitatively different about agents that you need some new permission level? Like a lot of people, okay, and I'll just kinda spew tokens at you until it like hopefully sparks something.Akshat [00:42:43]: Yeah.Swyx [00:42:44]: Like the new level now is whatever Claude Code does, which is dangerously scope permissions or like allow list by command or like whatever, right? And sometimes they're like, “Well, okay, we have like this adaptive thinking mode where like, just trust me, bro. I will make the calls for you.” Is that it? like mediated permissions.Hard Guardrails vs. LLM-Mediated PermissionsVibhu [00:43:03]: Now you're looping it with a goal and letting it roll.Akshat [00:43:06]: Yeah, I'm, I'm skeptical of LLM media permission for stuff that is at the sandbox level because you do want hard boundaries.Swyx [00:43:16]: Yeah.Akshat [00:43:16]: Otherwise, someone can exfiltrate stuff.Swyx [00:43:20]: But likeAkshat [00:43:20]: YeahSwyx [00:43:20]: Maybe that's old school thinking. Maybe we're the dinosaurs.Swyx [00:43:23]: Maybe the AI OS or the LLM OS is really the kernel is a goddamn LLM.Swyx [00:43:30]: Like it makes you feel uncomfortable.Akshat [00:43:31]: Yeah, I'm, I'm toldSwyx [00:43:32]: But that's what trusting the LLM is. Like imagine a spherical cow perfect LLM.Akshat [00:43:36]: Right.Swyx [00:43:37]: That it.Akshat [00:43:39]: Maybe.Swyx [00:43:41]: I wanna test the boundaries, right?Akshat [00:43:42]: Yeah.Swyx [00:43:42]: Like, and I don't believe that, but I wanna see where I'm wrong ‘cause that's, that's the consensus.Akshat [00:43:49]: Yeah. I think you always need hard guardrails when you want, And you can pair those with softer guardrails, right? And that's gonna be a lot of mediated.Managed Agents and Specialized SandboxesSwyx [00:44:00]: There. I'll also get you a end with a couple of your commentary on like the ecosystem outside of Modal. Manage agents. Everyone has one. Gemini, OpenAI, Claude, very useful for you, but also like it is their way of starting to edge into your space.Akshat [00:44:17]: Yeah.Swyx [00:44:17]: What's going on?Akshat [00:44:19]: Yeah, we're, very excited to partner with Anthropic and some of the other foundation labs, will not name who we're also working with. the way we see it is the manage agent thing is a great place to start if you're starting out building an agent and, But then when you get to, building something more production grade, like you're a company that's like Ramp that's building their own, Ramp also runs their accounting agent on us, so their external-facing agent. You need a lot more control over, your compute primitive on things like, what sort - how do you persist different files that the agent has access to, and how do you snapshot and restore? How do you control the networking? maybe you want GPUs. When you get to that point, you kinda want, a specialized sandbox provider, that gives you those things, and that's the role that we are trying to play.Swyx [00:45:15]: YeahAkshat [00:45:16]: We don't really have an opinion on the harness, whether it runs - it's a cloud-managed agent, and you hook it up to Model Sandbox, or you run the harness in Model Sandbox. We'll see where people converge with that.Swyx [00:45:26]: Yeah. Do you any opinions on like the meta harnesses, or just another layer on top of these things?Akshat [00:45:31]: You mean like the OpenPipeSwyx [00:45:33]: OpenPipe is one. I think Vercel had one, which I can't remember the name of right now. Fredshot had one. and then, to me, most recently was Data Databricks that had Omnigen. All these are meta harness. Like it's kinda pseudo agent cloud type things.Akshat [00:45:50]: I personally have not played around with them.Swyx [00:45:53]: Yeah.Akshat [00:45:53]: Build agents with them.Swyx [00:45:54]: Everything's bullish Modal, as long as it consumes more infra.Akshat [00:45:57]: That's why we're focusing on the infra layer. It's somewhere where our, relative competence is and, also it's a hard problem to solve.Swyx [00:46:06]: Yeah. I will say like just generally reflecting on that, I don't know if - if there's other topics on Modal, but like just generally reflecting as an infra person, not as intense as you, but in that field, this has like been the most exciting time in infra. Like it was boring for a while, and you couldn't really get people excited about data infrastructure. Like Eric would get on Data Console, everyone just watched the video and like say, “Look at how many sandboxes I can spin up,” and no one gave a crap.Why Infrastructure Became Exciting AgainAkshat [00:46:39]: Yeah.Swyx [00:46:40]: And like now everyone gives a crap.Akshat [00:46:42]: That's true. It is a very exciting time, and I think a lot of that's driven by just the amount of scale all of this stuff needs.Swyx [00:46:50]: I think the, like a lot of your initiatives or a lot of your like product directions make sense in retrospect, which is like the best kind, but I wouldn't necessarily have thought about it myself, which.Akshat [00:47:00]: We need the predictions.Swyx [00:47:02]: I think there's a lot that you just don't even see, right? Like you have the batch, you have the voice, you have the multimodal, but what else?Akshat [00:47:10]: What else is coming up for usSwyx [00:47:11]: Yeah. Where do you see things going?Akshat [00:47:13]: Yeah. I, in generalBiotech, Robotics, and Non-LLM AI WorkloadsAkshat [00:47:15]: It's it's clear that there's there's a huge shift happening. I think one thing that's not as obvious to people because LLM inference gets talked about so much and is also we work a lot of companies that are, doing things like drug discovery and computational bio, like the Chai Discoveries of the world. Big things are probably gonna happen there. we work a lot of robotics companies that are putting robots in like active deployments and getting good results out of them.Swyx [00:47:45]: Is there Air Gap Modal? Is there a version that is like prem air gapped whatever?Akshat [00:47:50]: No. We,Swyx [00:47:51]: You should cloud only.Akshat [00:47:51]: Yeah.Swyx [00:47:52]: Yeah. Okay. But yeah, so what you're saying is like because you're focused on primitives and they're good primitives, you find use cases in all these kinds of things.Akshat [00:48:01]: Yeah.Swyx [00:48:01]: Probably diversifies you a little bit away from LMS all the time.Akshat [00:48:05]: Yeah, absolutely. We're, we'- our goal isn't to only serve the LLM inference market.Swyx [00:48:10]: There are a lot just on the website, the audio,Akshat [00:48:12]: Yeah. We said both onSwyx [00:48:14]: Computational bio images. Yeah, there's a lot here. There's QTA TTS, customizing. Oh, Chatterbox. there was customizing Whisper.Akshat [00:48:24]: Okay. Yeah.Swyx [00:48:25]: This screen reminds me of a fallen competitor, which Replicate.Model APIs vs. Differentiated AI ProductsSwyx [00:48:31]: What's your postmortem on what happened?Akshat [00:48:34]: This is one thing we've stayed away from is providing an API for models because I think providing model APIs is some of it ends up serving like a really hobbyist market, which is much less sticky.Swyx [00:48:50]: Yeah.Akshat [00:48:50]: And we've always wanted to build for companies that are building products and need more flexibility that's not just an API.Swyx [00:48:57]: Which you can build an API for a model and this is clearly what it is. But you - but what you're saying, you can wrap it into a more fully functioning back end that you run.Akshat [00:49:06]: Yeah. So all of our examples, it's not that spin up this model, here's an API token, use it. They're all code.Swyx [00:49:13]: Okay.Akshat [00:49:13]: And so the point is that this is just an example.Swyx [00:49:16]: Starter code.Akshat [00:49:17]: Yeah. But you can tweak it however you want.Swyx [00:49:20]: Yeah.Akshat [00:49:21]: And if you're like a company building a product, like, computational bio whatnot, yeah.Swyx [00:49:26]: I guess I'm trying to tease out for listenersAkshat [00:49:28]: YeahSwyx [00:49:28]: When does it stop becoming, oh, you're just an API call and you're just a wrapper on API to becoming what you call a product, right?Swyx [00:49:36]: Like, what is that layer? Like what-- Like, more lines of code, but like beyond that, what is the substance that people add that qualifies it to be something more?Akshat [00:49:46]: I think there's a little bit of like a selection effect of like a lot of the companies who do wanna get deeper into that level are probably building something that's more differentiated. And, I think, an example is like - with LLM inference, originally we, worked with companies that were building their own post-training frameworks or they were, - Ramp early in the day was training their own tokenizer and like swapping out the tokenizer in Llama and whatnot. I'm not saying that's, that successful, in that case. But a better example is like, let's say Suno. because Suno, does not use Modal for training.Swyx [00:50:26]: Mikey on the pod. Yeah.Akshat [00:50:27]: But they use Modal for all their inference and that's because they have like a custom-- They have completely custom model architecture and that means that they have to be at the code level and tweak things that are not, just an API.Swyx [00:50:41]: It's interesting as well, like we had, Ethan, most recently on the xAI Groq team make a prediction that like the next tier in video gen is not a better video model, it's a better model or agent that orchestrates video models.Video Agents and Production WorkflowsAkshat [00:50:56]: Oh, interesting.Vibhu [00:50:56]: Language model backbone that can use toolsAkshat [00:50:58]: RightVibhu [00:50:59]: And write code.Akshat [00:51:00]: Like, yes, I can make my second video or my second video from Groq, but I want my minute video.Akshat [00:51:06]: And I'm not going there through normal video gen.Swyx [00:51:10]: Yeah, that's interesting. I - So we have GPU sandboxes and recently have seen a few companies doing agents that do video manipulation or,Akshat [00:51:22]: Yeah. Give it FFmpeg and just do it.Swyx [00:51:23]: Run FFmpeg. But likeAkshat [00:51:25]: That's not enough.Swyx [00:51:25]: Yeah.Akshat [00:51:26]: You need to give it Adobe.Swyx [00:51:27]: Yeah, I hadn't put it together with like it would be a video production thing. in my mind these things were going more towards editingAkshat [00:51:36]: Yeah.Vibhu [00:51:36]: Well, shout out Mantis.Akshat [00:51:37]: I think about this a lot.Swyx [00:51:38]: .Akshat [00:51:41]: Yeah. Sorry.Vibhu [00:51:41]: Luma. Luma Agent is a version of this for video production, but it's a off.Swyx [00:51:46]: I was gonna get your quick takes, on some other stuff that happensGitpod/Ona, CI, and Runtime SandboxesSwyx [00:51:50]: In recent news and just-just see if you have anything interesting. Gitpod, very li

Join us as Du'An digs into the real mechanics of running AI locally and in production - from GPU memory math to multi-agent architectures, observability, and the economics of self-hosted inference. Du'An walks through how model weights and KV cache compete for GPU memory, why continuous batching matters when you have more than a handful of users, and how agent architectures like single-agent, workflow, graph, swarm, and supervisor patterns each solve different problems. You will learn how to instrument your agents with Langfuse for observability and cost tracking, when to use Ollama versus vLLM, how prompt caching can cut provider costs by up to 75%, and why GPUs should never sit idle. Episode two of three - the next episode covers deploying at scale. Timestamps 0:00 Welcome & Introduction 1:47 Du'An's New Role at Akamai Cloud 3:10 Data Privacy and the Case for Self-Hosted AI 7:21 Anthropic and OpenAI as the New Cloud Layer 12:48 Local Models for Specific Use Cases - Cancer Detection Example 15:02 GPU Memory Math - Weights, KV Cache, and Context Windows 19:32 Continuous Batching and GPU Time Slicing 20:03 Observability with Langfuse - Live Demo 27:44 Agent Architectures - Single Agent, Workflow, Graph, Swarm, Supervisor 36:36 Token Economics, Prompt Caching, and GPU Cost Planning 45:32 Ollama vs vLLM - Prototyping vs Production How to find Du'An: https://duanlightfoot.com https://www.linkedin.com/in/duanlightfoot/ Links from the show: https://langfuse.com/ https://github.com/akamai-developers/akamai-workshop-solution-architect-agent https://amzn.to/4bvHn1p https://vllm.ai/

Leif Nilsson, CEO & Director of Surge Copper (TSX.V:SURG – OTCQB:SRGXF), joins me for a comprehensive update covering the updated Mineral Resource Estimate and Pre-Feasibility Study (PFS), at their flagship copper-molybdenum-silver-gold Berg Project in British Columbia.   Leif mentioned that the completion of the Berg PFS marks an important milestone for Surge and materially advances one of Canada's most significant undeveloped copper projects. Berg now stands out not only for its scale, but also for the quality of its development profile, with long-life production of copper as a primary metal, and industry leading molybdenum and silver output, strong infrastructure advantages, and access to low-carbon hydroelectric power. Just as importantly, this study reflects a great deal of technical work completed since the PEA and provides a more defined foundation for the next stage of advancement, including continued work with First Nations, formal entry into the environmental assessment process, and future feasibility-level studies.   Key highlights from PFS:    Base case after-tax NPV8% of C$4.6 billion, IRR of 24%, and payback period of 2.9 years, based on long-term commodity price assumptions of US$4.75/lb copper, US$20.00/lb molybdenum, US$45/oz silver, and US$3,500/oz gold and an exchange rate of 0.73 US$/C$ At spot prices as of June 2026 (US$6.45/lb copper, US$30.00/lb molybdenum, US$65/oz silver, and US$4,250/oz gold and an exchange rate of 0.73 US$/C$), a spot price sensitivity case generates an after-tax NPV8% of C$9.4 billion, an IRR of 36%, and a payback period of 1.8 years, underscoring the Project's leverage to higher metal prices Maiden Proven & Probable Mineral Reserve of 1.2 billion tonnes grading 0.22% copper, 0.026% molybdenum, 4.1 g/t silver, and 0.02 g/t gold, containing 5.8 billion pounds of copper, 687 million pounds of molybdenum, 160 million ounces of silver, and 0.8 million ounces of gold Updated Mineral Resource Estimate includes Measured and Indicated Mineral Resources of 1.4 billion tonnes grading 0.21% copper, 0.025% molybdenum, 4.0 g/t silver, and 0.02 g/t gold, plus additional Inferred Mineral Resources of 1.0 billion tonnes grading 0.16% copper, 0.027% molybdenum, 4.3 g/t silver, and 0.01 g/t gold 28-year mine life with total production of 8.6 billion pounds (3.9 million tonnes) of copper equivalent (CuEq)1, including 4.9 billion pounds (2.2 million tonnes) of copper, 602 million pounds of molybdenum, and 89 million ounces of silver First 5 years of steady-state production averages 416 million pounds (189 thousand tonnes) of copper equivalent annually, including 270 million pounds (122 thousand tonnes) of copper, 21 million pounds of molybdenum, and 4 million ounces of silver Life of mine average annual production of 308 million pounds (140 thousand tonnes) of copper equivalent, including 176 million pounds (80 thousand tonnes) of copper, 21 million pounds of molybdenum, and 3 million ounces of silver Life of mine C1 co-product cash costs of US$1.95/lb payable CuEq and by-product cash costs of US$(0.17)/lb payable Cu Low life of mine strip ratio of 2.0:1, including waste pre-stripping requirements of 304 million tonnes Initial capital cost of C$4.7 billion and sustaining capital of C$1.7 billion, based on an EPCM execution approach and a three-year construction period, and including a total life of mine contingency of C$715 million, implying initial capital intensity of US$24,416/t CuEq annual average production capacity, and life of mine capital intensity of US$0.55/lb recovered CuEq Selected development case based on a 120,000 tonne per day concentrator and a new 230 kV transmission line connecting the Project to the BC Hydro grid, and downhill overland conveyor transport of ore to the process plant Simple, stand-alone development case based on a single-phase build, conventional open pit mining and processing, with no reliance on phased expansions or third-party major infrastructure     If you have any follow-up questions for Leif regarding Surge Copper, then please email them to me at Shad@kereport.com.   In full disclosure, Shad is a shareholder of Surge Copper at the time of this recording, and may choose to buy or sell shares at any time.   Click here to follow the latest news from Surge Copper   For more market commentary & interview summaries, subscribe to our Substacks:   The KE Report: https://kereport.substack.com/ Shad's resource market commentary: https://excelsiorprosperity.substack.com/     Investment disclaimer: This content is for informational and educational purposes only and does not constitute investment advice, an offer, or a solicitation to buy or sell any security. Investing in equities and commodities involves risk, including the possible loss of principal. Do your own research and consult a licensed financial advisor before making any investment decisions. Guests and hosts may own shares in companies mentioned.    

I dagens avsnitt reser vi ända till år noll och landar utanför ett litet stall i Betlehem en sval, stjärnklar natt! Kvällens förtrollande saga "Edda och den gömda blomman" är önskad av Edda, 7 år från Teckomatorp.Inuti ett gammalt, knotigt träd hittar Edda en talande, magisk blomma som heter Luma – och får plötsligt grön växtmagi i händerna. Genom en virvel av ljus dras hon till Grönskedalen, där den kalla ishäxan Mirea håller på att frysa allt levande till is. Edda måste våga stoppa henne, och upptäcker att även en häxa kanske bara längtar efter något som får stanna kvar.I studion hjälper gänget trötta herdar att tysta en flock bräkande får så att bebisen i krubban får sova, medan Aida får en liten existentiell kris över att "år noll" faktiskt inte finns i kalendern. Plus fem fakta om vardagslivet för 2 000 år sedan.Stötta podden och få tillgång till nya sagor! Gå med i Magiska Godnattsagor-klubben!Skicka in förslag på kommande sagor via www.magiskagodnattsagor.seSökord: magiska godnattsagor, godnattsaga, barn, läggdags, podcast för barn, barnlitteratur, ai, godnatt, betlehem, växtmagi, ishäxa, mod, vänskap, Teckomatorp

Kväänifästistä, alkuperäskansa-staatyksestä ja markkinoitten valmistelutöistä Pajalassa. Lyssna på alla avsnitt i Sveriges Radios app.

Brankář Antonín Kinský je jednou z nejpozoruhodnějších postav tuzemského profesionálního sportu, která v době reprezentačního zmaru ukazuje úspěšnou tvář českého fotbalu. Vždyť tři roky nazpět nastupoval za druholigový Vyškov a teď podepsal nový kontrakt s londýnským Tottenhamem za desítky milionů korun. Ovšem ještě na přelomu jara a zimy byl tamtéž na odpis a pomýšlel na odchod. Jeho příběh je zkrátka plný pádů a vzestupů. „Vždycky jsem měl jasný cíl a disciplínu,“ shrnuje svou cestu třiadvacetiletý gólman v podcastu Money Maker.V rozhovoru se dále dozvíte:

Karlovarský filmový festival slaví 60. narozeniny a chystají se tam opět davy lidí. Seriál Radiožurnálu se ale vrací do doby, kdy filmové přehlídce vládla ideologie a kina byla poloprázdná. „Po roce 1968 následovala pauza a v roce 1970 na 17. ročníku už byl cítit závan normalizace,“ vypráví kronikář festivalu Květoslav Kroča. Systém hodnocení filmů se vrátil do 50. let. V roce 1978 se festival poprvé konal v nově otevřeném hotelu Thermal. Jak ho hodnotil Jiří Bartoška?Všechny díly podcastu Seriál Radiožurnálu můžete pohodlně poslouchat v mobilní aplikaci mujRozhlas pro Android a iOS nebo na webu mujRozhlas.cz.

I slutningen af 1970'erne rystes Los Angeles af en række brutale fund, hvor unge kvinder findes efterladt på byens øde bjergsider. Efter serien om Rodney Alcala, ”Den charmerende seriemorder”, vender vi nu tilbage til samme tid og sted for at undersøge sagen om Kvæleren fra bakkerne. Fortællingen går helt tæt på en efterforskning, hvor politiet er oppe imod en gerningsmand, der kynisk udnytter falske politiskilte og lovens egen autoritet til at vinde sine ofres tillid. Det er samtidig beretningen om en by i knæ, et komplekst politiarbejde og jagten på en morder, der gemmer sig lige for øjnene af alle.  Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

V tomhle díle jdeme do evoluční a kognitivní psychologie přes konkrétní experimenty, které dodnes nahání husí kůži. Poslušnost autoritě, misatribuce emocí, halo efekt, iluze kompetence i důvěra, která vzniká zvláštní kombinací přátelskosti a autority. Tohle je zkrácená verze (44min) poslouchej celý díl (76min) jen za 100kc/měsíc! Zde: https://open.spotify.com/episode/43sNP0ZiovypG1MP7wSTSD?si=d2aff5d4df564f14Je to epizoda o tom, jak vznikají naše omyly, proč jsou tak přesvědčivé a proč nestačí mít „správná data“, když je mozek nastavený rychle označkovat realitu a jít dál. A taky o tom, že perspektiva není měkké klišé. Je to tvrdý zásah do toho, co cítíš, jak jednáš a jaký svět se ti vůbec ukazuje.Parťáci epizody jsou:KusKakaa - ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠www.kuskakaa.cz⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠  Přináší do Česka čistá ceremoniální kakaa. A proč si takové kakao dopřát? Ukazuje se, že přináší celou řadu benefitů a má velký obsah flavonoidů a polyfenolů. Tak jdi na ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠www.kuskakaa.cz⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ a zkus jedno z jejich kvalitních kakaí! Doporučujeme to z Kostariky, nebo Peru.N⁠⁠⁠⁠⁠⁠orsan.cz⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Norsan Vyrabí Omega 3 z čerstvého rybího oleje z udržitelného rybolovu nebo z mořských mikrořas. Jdi na Norsan.cz zadej kód bwa10 pro 10% slevu a pořiď si kvalitní OMEGA-3 tvůj mozek a zdraví ti poděkuje.Minutáž:00:00 Úvod05:17 Milgramův experiment a vliv autority07:44 Fenomén šesti stupňů odloučení09:49 Fenomén šesti stupňů odloučení12:06 Hodinář a argument inteligentního stvořitele14:38 Aristoteles, Darwin a počátky evoluční myšlenky20:05 Evoluční důvody sexuální averze k blízkým21:24 Evoluční zpoždění u moderních hrozeb23:05 Model lidské motivace a pocit sounáležitosti25:47 Funkce emocí a sociální cíle27:16 Kognitivní podstata budování důvěry28:34 Misatribuce emocí a experiment na visutém mostě33:14 Valinsův experiment s falešným tlukotem srdce35:21 Falešná interpretace lásky v dramatických vztazích39:12 Pratfall efekt a sympatie z vlastních chyb40:49 Dopaminová křivka a efekt získávání sympatiíPřechod do VIP- Milgramův experiment a síla sociální konformity- Kognitivní kapitulace vůči umělé inteligenci- Nebezpečí slepé důvěry v chyby umělé inteligence- Agentní pareidolie a iluze umělého vědomí- Amnézie zpráv a ztráta kritického myšlení v neznámých tématech- Logická hádanka s pálkou a míčkem- Konformita a vliv skupiny na vnímání barev- Vznik konfliktů a usmíření na letním táboře- Haló efekt a fundamentální atribuční chyba- Matoušův efekt a prohlubování sociálních rozdílů- Kvízový experiment a falešné posuzování kompetence- Efekt blábolení a jeho využití v politice- Síla přerámování vnímání reality a vědomé změny prožitku- Buddhistický příběh o voru a opouštění zátěže- Indické podobenství o slepcích a slonovi- Závěrečný citát o inteligenci a lidském charakteru

Två dagar efter får Felix syn på två av de män som slog ner honom. Nu står han blåslagen framför dem. Vad ska han göra? Konfrontera eller förlåta? Lyssna på alla avsnitt i Sveriges Radios app. Reporter & ljuddesign: Leontine OlsbjörkProducent: Gustav AsplundSlutmix: Astrid AnkarcronaVerkligheten görs av produktionsbolaget Filt.UTSKRIFT AV AVSNITTET:– Och jag kände först en ångest av att gå ut för att jag. Ja, jag har precis varit med om någonting väldigt traumatiskt och jag ser ut som jag gör. Men tänker väl ändå att det här ska jag ska inte skämmas eller jag ska väl inte behöva stanna inne för att jag har varit med om det här.Felix är blåslagen i ansiktet. Två dagar tidigare har han blivit misshandlad på en nattbuss.När han sa ifrån mot några unga män som uttryckte sig rasistiskt.– Jag står där och värmer mig vid en eld och ser då två av de här. Killarna kommer gåendes.jag säger då till min familj att där är de ju.Nu står han mitt emot två av dom unga män som misshandlade honom. Nu, ska han konfrontera killarna – och förlåta dem.***Första gången jag ser Felix är i TV4 Nyhetsmorgon. Jag står i green room och pillar nervöst på en kaffekopp. Om en liten stund ska jag intervjuas i direktsändning.Du kanske har hört min berättelse. För Verkligheten har jag berättat om när jag tog ett DNA-test och hittade över 30 halvsyskon. Och den här morgonen ska jag prata om det i Nyhetsmorgon. Det är december och i sminket pratas det om det stundande luciatåget. Framför mig sitter flera tv-skärmar uppsatta på väggen. I dem ser jag Felix.Han sitter i en soffa inne i studion. Drygt två veckor tidigare har han blivit misshandlad på en nattbuss, efter att ha sagt ifrån mot några unga män som uttryckt sig rasistiskt. Jag har fått pushnotiser om misshandeln tidigare, men nu ser jag Felix berätta om kvällen. Fortfarande med blåmärken under ögonen. Och det är en sak jag fastnar för. För Felix säger att han har förlåtit dem.Intervjun tar slut och innan Felix försvinner vidare genom korridoren och jag in i studionhinner vi hälsa snabbt. Men det är något som jag inte riktigt kan släppa.Några månader senare sitter jag på ett tidigt morgontåg norrut, mot Vännäs i Västerbotten. Jag ska träffa Felix, igen. Felix möter mig upp vid tågstationen– Här har vi Myranparken.Vi går genom den lilla bruksorten några mil utanför Umeå.– Min farfar bodde i det här gula huset där. Då såg parken lite annorlunda ut när vi var små.Felix och hans syskon brukade leka här som barn.– Det var ganska mycket mer växtlighet här, så man kunde klättra i några träd-ish buskar och sådär. Och så har man ett starkt minne av en rutschkana som var lite farlig, men det är lite andra regler idag.Han växer upp i en annan liten ort, Rundvik, men han var ofta i Vännäs som barn. Och som vuxen flyttar han hit.– Pappa var väl inget stort fan av Vännäs. Han sa till mig innan han gick bort till 2020. Han sa att du får fan inte flytta till Vännäs. Men då träffade jag en Vännäs-bo. Så han får vända sig graven om han vill.När Felix är liten separerar hans föräldrar. Hans pappa missbrukar och kontakten är sporadisk.– Han var också en klok man. Var också närvarande när vi väl var där. Mycket starka minnen om att vi hittade på saker. Så man såg upp det handeln då. Fick mycket från honom i just hur jag är som person. Väldigt, väldigt... Jag analyserar mycket, jag tänker, reflekterar. Och det har inte alltid heller varit lätt.Felix mamma är från Brasilien och familjen är den första med invandrarbakgrund i byn.– Man fick höra mycket när man växte upp. Väldigt mycket glåpord. Och i och med att min familj inte jobbade heller och skaffade många barn så var man ganska utsatt på grund av det. Det var många som visste vilka vi var utan att man hade pratat med dem. Och mycket fördomar.De har det tufft ekonomiskt. Felix och hans syskon går ofta klädda i ärvda kläder och det händer att dom går i fel kläder för säsong. De sticker ut.– ´Det var en inre stress hela tiden. Man skämdes för att vi hade så mycket syskon som levde om och inte kunde sköta sig. Enligt en själv då. Idag var ni ju barn så det är svårt att klandra dem. Men det var så man tänkte då i alla fall.På skolgården fortsätter fortsätter glåporden.– Men det var ju mycket på skolan. Man fick ju höra diverse. Allt från svartskalle till turkjävel.Ibland säger han ifrån och andra gånger skrattar han bara med– Men många gånger svalde man det också. Det var som att bita ihop och hantera det. Jag har svårt att använda ordet rasism. Jag vet inte varför. Men det var helt klart rasistiskt och det var helt klart okunskap.Felix känner sig utanför. I ett försök att passa in bleker han sitt mörkbruna hår.– Det slutade inte så bra för att det blev orange. Det var ytterligare en sak som var såsynlig att man gjorde ett försök som inte slutade bra.Hemma får Felix och hans syskon höra att det är viktigt att de sköter sig i skolan och bland vuxna.– Jag var väldigt mån om att följa regler och hjälpa till där det behövdes städa undan och respektera de äldre. Upplevelserna när jag var liten, det har ju absolut format mig idag. Jag har noll tolerans emot mobbning generellMen skolan var också en plats där Felix blev sedd och även om det stundtals var tufft hemma,så kommer han från en kärleksfull familj. Vi spolar fram till en fredagskväll i november förra året, då utanförskapet och hatet gör sig påmint igen. För Felix växer upp. Han är social och utåtriktad. Han brinner för barn och unga, jobbar på fritidsgård och som ungdomstränare. Han börjar plugga till lärare med hoppar av och börjar jobba inom skogsindustrin.– Jag hade precis varit på en AW.På en arkadhall i Umeå står Felix och hans kollegor bland blinkande pinball-maskiner och gamla retrospel.– Så jag hade bokat det och så hade jag fixat pizzor till alla och så där. Så vi hade varit där hela kvällen och umgåtts och haft jättekul.Det är en trevlig kväll. Men så börjar samtalen förändras.– En sak som som kom upp i samtalet där under kvällen var just politik. Där jag blev lite förvånad över folks åsikter och värderingar och så där. Men vi valde sedan att gå vidare då.AW:n fortsätter. Felix är på gott humör när han skiljs från sina kollegor. Han ska ta nattbussenhem till Vännäs och beger sig till busshållplatsen.– Jag hade börjat prata med två med två kvinnor där på busshållplatsen och skrattat och skämtat och haft kul. Men det där låg lite grann i bakhuvudet också, just över den politiska den diskussionen vi hade Det var väl därför jag kanske också initierade i det här samtalet med de här kvinnorna.Kvällens samtal gnager fortfarande i honom när han börjar prata med kvinnorna. De går på bussenoch sätter sig långt fram där fortsätter samtalen.– De sätter sig ganska nära mig så att jag vänder mig och börjar prata med de här två kvinnorna. Och ja, vi pratar om allt möjligt och jag frågar hur länge de har bott här i Vännäs och hur de ser på på klimatet här i kommunen. Vi delar lite åsikter och tankar och sådär då. Och vi sitter och pratar egentligen hela vägen från från Umeå till Vännäs by där de hoppar av.Men när kvinnorna kliver av bussen förändras stämningen snabbt Felix hör några killar längre bak i bussen prata han hör att de säger något som slår an i honom. De säger n-ordet flera gånger– Så jag går bak och sätter mig och sätter mig och sätter mig och sätter mig framför dem och hälsar och säger såhär till grabbar. Så här kan ni inte uttrycka er. Ni måste tänka er för på hur ni pratar och hur ni uttrycker det, för det här är inte okej. Jag blir väl bemött ganska ignorant, de tycker att de får säga vad de vill och de har inte sagt någonting som inte är lämpligt.Några säten längre fram sitter en kvinna och lyssnar. Hon vänder sig om.– Och också säger till den här kvinnan är ju då svart. Så hon har ju slutat höra på allt det här vad de har sagt, så hon vänder sig om också och säger, säger åt dem. En av de här personerna blir då ganska, störig. Han tycker, han försöker liksom provocera mig och så där. Och så efter lite, lite fram och tillbaka så är vi då framme i Vännäs näst sista hållplatsen och då säger de ”ska vi kliva av här”?Varför jag svarar ja, det kan vi göra.Felix reser sig upp och går mot bussens utgång.– Men då upptäcker jag att jag har glömt min väska på mitt säte.Han vänder tillbaka in genom bussen– Var på den här personen skriker “vad fan gör du” och tar stryptag på mig. Och jag ramlar som över honom där han satt och då slår han mig två gånger. Felix blir slagen med knytnävsslag i ansiktet. Han försöker hålla fast killen för att avvärja slagen och skydda sig. – Och jag säger till honom att att är det bra nu? Var på det någon bakifrån börjar slå mig. Och då tar jag tag i honom också och håller fast de två. Och så började en tredje slå mig. Ingen säger någonting. Jag blir slagen, slagen, slagen.Killarna fortsätter slå Felix medan bussen åker vidare mot ändhållplatsen– När bussen stannar så släpper jag sista personen som jag håller i. Killarna försvinner iväg. Felix halvligger över sätena längst bak i bussen. Ingen på bussen hjälper Felix. Och jag sätter mig upp och känner hur det rinner blod från ansiktet. Så till slut så ställer jag mig upp och tittar upp och då ser jag busschauffören stå där och han säger ingenting. Så jag kliver av bussen och bussen kör iväg.Skärrad och blodig står Felix på busshållplatsen mitt i natten. Han plockar upp sin mobil och slår numret till 112. Efter en stund kommer både polis och ambulans, polisen ställer frågor och ambulanspersonalen undersöker Felixs ansikte och hals. Han förs till akuten i Umeå. Samtidigt försöker Felix på tag på sin sambo, men hon svarar inte. – Så jag skickade en bild i en familjegrupp vi har. Och förklarade vad som hade hänt lite snabbt. och att de måste försöka få tag på henne. På bilden är hans högra öga igensvullet. Ansiktet är blodigt och rödflammigt.– Jag grät och jag fick panik. Hur ska jag berätta det här för min sambo? Mina barn ska se mig också.Felix har fått flera frakturer i ansiktet, i kindbenen och näsan. Först framåt morgonen får han tag på sin sambo.– Jag kommer ihåg att jag sa till henne att jag var på en säker plats och ändå mådde bra. Det gör det jobbigt nu också när jag säger det högt att jag inte kommer ihåg hur hon reagera.samtidigt går tankarna på högvarv.– Jag var också så chockad över att det var ett sådant våldskapital. Det är inte första gången jag sitter på den här bussen hem efter en utekväll och har haft samtal med diverse personer och stoppat bråk mellan andra personer. Jag brukar kunna se var saker och ting är på väg. Här hade jag ingen aning. Jag förstod mig inte alls. Jag kunde inte se det framför mig. Jag kände ingenting som att det skulle bli sådant här våld. och så tyckte jag synd om dem för just av den anledning, att det va var verkligen så oväntat så det gjorde ont i mig att unga människor är så arga.Det har gått två dagar sedan misshandeln. Det är skyltsöndag. Byn är full av människor. Det är ponnyridning, marknad och tomtebesök.– Jag kände först en ångest av att gå ut. Jag har precis varit med om något väldigt traumatiskt och jag ser ut som jag gör. Jag tänker att jag inte ska skämmas för att jag har varit med om det här.Trots att han är synligt blåslagen i ansiktet följer Felix med sin familj till byns torg.– Jag står där och värmer mig vid en eld och står och pratar med min mamma och min syster. Då ser jag två av de här killarna komma gående. De ser att jag ser dem och då vänder de ganska fort. Jag säger till min familj att där är de. Mamma säger åt mig att inte följa efter.Men Felix följer efter dem. De skyndar iväg till en parkerad bil.– De hinner sätta sig i bilen men jag går fram och knackar på förarfönstret. Jag sätter mig på huk och hälsar och säger hej känner ni igen mig Killarna nekar. Och Felix blir allt mer irriterad.– Så ni känner inte igen mig? Från bussen i fredags. Då säger de att, nej, vilken buss? Det här är samma attityd ni hade då som nu. Det är inte okej. Okej, ja, men hur ska vi lösa det här nu då? Jag säger något i stil att de ska se på mig och titta vad de har gjort med mig. Då mjuknar de lite och ber dem ursäkt för det de gjorde. Då säger jag till dem att de gärna får be sina kompisar ringa mig också. Jag lämnar mitt nummer och mitt namn. Sen kör de därifrån.Men Felix känner sig inte klar. Han söker på bilens registreringsnummer. han får fram adressen, men hejdar sig.– Jag kände att jag inte kan träffa honom eller dem ensamma. Det kändes inte bra. Istället tar han reda på var hans föräldrar bor.– Jag åker dit och parkerar bilen. Jag tänker nog inte så mycket på vad jag ska säga annat än att presentera mig och berätta vad som har hänt. Jag knackar på dörren och så öppnar en man.Det visar sig att mannen som öppnar är styvfar till en av killarna i bilen– Jag hälsar och säger hej. Jag heter Felix. Jag berättar att jag har blivit misshandlad på bussen denna fredags. Jag frågar om jag får komma in. Han bjuder in mig och vi sätter oss i deras vardagsrum.Det får han. Styvpappan bjuder felix på en kopp kaffe, och de sätter sig ned i vardagsrummetoch pratar.– Och jag säger väl att jag inte vet hur jag ska hantera det här på något annat sätt än vad jag gör just nu. Jag frågar om det var konstigt att jag var där och han tyckte att det inte var det. Vilket jag har varit tacksam för.En stund senare kommer mamman till en av killarna hem.– Hon frågar mig då om hon ska ringa hem grabbarna igen då. Så de kommer dit också.Nu möter Felix två av killarna igen. De sätter sig ned och pratar.– Jag ville visa att så här ser jag ut idag. Jag åkte dit av känslan att. Jag tänker inte gömma mig för det här. Utan personerna måste verkligen få se vad de har gjort och vad... Det finns en person på andra sidan här också. Och att saker får konsekvenser.Killarna ber om ursäkt igen och den här gången svarar han att han förlåter dem– Jag känner väl en sorg över att, att de valde det här, för det är också någonting de gjorde.Men att de inte har haft samma möjligheter kanske eller samma förutsättningar för att kunna reglera sig själv eller hantera sig själv. Eller fått stöd och hjälp i just känslor.– För jag har också varit i den åldern och jag var också ganska arg. Men jag var inte arg på... Men jag var mer arg på var jag kommer ifrån. Jag kommer själv från förhållanden som... Som har varit långt ifrån optimala. Jag kommer från ett kärleksfullt hem, absolut.Men jag kommer ifrån mycket och stora utmaningar.Felixs val att förlåta killarna tas emot med blandande känslor och en vän till honom är orolig att han inte kommer medverka i polisutredningen.– Och han har också sagt till mig rakt upp och ner. Att... Varför ska du vara så här för? Varför ska du... Varför ska du ens förlåta dem? Du är inte bättre än oss andra. Han har sagt. Och det är jag inte. Inte alls. Men sen kanske man också måste tillåta sig själv att få vara. Arg då eller ledsen eller sårad eller utsatt.Och såhär ett halvår senare märker han att misshandeln, precis som uppväxten, har satt sina spår– Och jag har ju fått frågan tidigare om jag kommer fortsätta ingripa och säga ifrån. Och jag har väldigt tryggt sagt ja, det här kommer inte förändra mig. Och ja, det har det gjort. Jag är mer osäker nu än tidigare. Att ha allt det där i ryggsäcken och så får man växa upp i en mindre ort med mycket fördomar och rasism. Det skapar ett annat förhållningssätt till livet än vad många här i Sverige kanske förstår.– Jag är supertacksam för det Sverige jag har växt upp i. Hela skyddsnätet som vi har haft är därför jag sitter här idag. Jag hade kunnat vara på en helt annan plats om jag inte hade haft det. Det måste vi värna om, att vi tar hand om våra medmänniskor som inte har allt på det torra.Efter misshandeln samlas människor i Vännäs i en manifestation mot våld och rasism. När de tågar genom byn hörs ropen: “Alla hjärtan har samma färg.”– All den kärlek och värme jag fick från hela Sverige, från olika människor. Det var personer som skrev brev till mig och personer som hörde av sig på alla möjliga olika medier. Och det visar ju på att det finns en kraft och att det finns människor där ute som på riktigt bryr sig. Och det är lätt man glömmer det, tänker jag. Det är lätt att man glömmer bort att det här berör så många. Det har jag inte tänkt på länge nu så det kanske jag ska försöka tänka på oftare. Just att den värmen och kärlek jag fick efter allt det här. Jag ska med mig det och att det kanske också kan få andra att agera i framtiden också. Om de ser eller hör eller upplever någonting.

Postaral sa o jeden z najdiskutovanejších a najpikantnejších prestupov na domácej scéne MMA. Stálica Oktagonu, víťaz Výzvy a bývalý dočasný šampión Samuel „Pirát“ Krištofič zakotvil v RFA. V podcaste Staredown na ŠPORT.sk otvorene prehovoril o pocite dešpektu zo strany promotéra Oktagonu Ondřeja Novotného, o ponukách z iných organizácií, o pokuse Oktagonu na poslednú chvíľu udržať si ho, no aj o tom, či má v sebe ešte oheň a prečo už najbližší zápas môže byť posledný.Čo sa dozviete v rozhovore?Kvôli čomu sa definitívne rozhodol skončiť v Oktagone?Čím konkrétne mu Ondřej Novotný neprejavil rešpekt?Prečo jeho názor nezmenila ani ponuka na odvetu s Vémolom?Má v sebe ešte oheň alebo ide do RFA len kvôli peniazom?Prečo označil slovenský prospekt Martina Kozáka za hlúpeho?Nakoľko reálny je zápas s Machom Muradovom v RFA?Koľko zápasov ešte v sebe má a prečo už najbližší môže byť posledný?Prijal by ponuku Clashu a ďalších bizarných organizácií?

Kvällspasset är på plats hemma hos Karl Fredrik Gustafsson på Österlen för att prata midsommarminnen, blommor, mat och musik! Lyssna på alla avsnitt i Sveriges Radios app. Ett nyfiket och underhållande aktualitetsprogram med lyssnaren i fokus.Tillsammans med floristen och tv-profilen Karl Fredrik, artisten Diane Emerita, matinfluencern Julia Tuvesson och lyssnarna tar vi oss an midsommarafton.Julia bjuder på midsommarpizza, Karl Fredrik fixar kransarna och Diane sjunger bland annat en vacker tolkning av Lasse Berghagens ”En kväll juni”.

In this episode of The Right Idea, Derek Cohen sits down with Carson Clayton, TPPF Life Powered Campaign Director, to break down the controversial 765 KV transmission lines (Strategic Transmission Expansion Plan / STEP).Texas lawmakers and activists are sounding the alarm over a $33 billion plan to build massive ultra-high voltage transmission lines across the state — cutting through pristine Hill Country and farmland — instead of addressing the root cause: Texas' broken energy market that over-subsidizes intermittent wind and solar while under-building reliable dispatchable power.Featuring powerful clips from Senator Kevin Sparks and Representative Brad Buckley.Key Topics:Why the Permian Basin — one of the most energy-rich regions in the world — needs power imported from Central TexasHow federal subsidies, ESG pressure, and ERCOT's energy-only market are distorting investmentThe massive cost to ratepayers and landownersWhat real market reform looks like to prevent blackouts and unnecessary transmission boondogglesTimestamps:00:00 - Welcome & Introduction to the 765 Lines Controversy01:23 - What Are the 765 KV Transmission Lines?02:41 - Senator Kevin Sparks on the Permian Basin Plan04:40 - Why Wind & Solar Boom Created a Reliability Crisis08:37 - Senator Sparks on ERCOT Market Failures & Subsidies09:39 - How the Energy-Only Market Rewards Unreliable Power12:19 - Federal Policy, ESG, and Renewable Credits Driving the Problem15:12 - Rep. Brad Buckley: Pause the Project & Reform the Market17:02 - Proposed Market Reforms (SB 715 & Reliability Standards)20:01 - The Coming Reliability Cliff & Why Transmission Is Just a Band-Aid21:22 - How Much Gas Generation Would Make the 765 Lines Unnecessary?If you care about Texas energy independence, affordable electricity, property rights, and keeping the lights on, this is a must-watch.

Narodil se 17. června 1909 v Brně jako 13. dítě. Měl silné pouto k mamince. I kvůli ní se stal učitelem. Láska k herectví ale zvítězila. V roce 1940 odešel do Prahy do Národního divadla, kterému už zůstal věrný. Zachoval si ale svůj charakteristický měkký moravský přízvuk. Ztvárnil bezpočet rolí na divadle ve filmu i v rozhlase, kde se podílel i na vysílání v esperantu. Byl citlivý, přemýšlivý, empatický, uctivý k ženám. Kvůli pomoci druhým riskoval za války svůj život.

Západní balkon katedrály sv. Víta na Pražském hradě už pamatuje leccos – od středověkých požárů přes bombardování Prahy až po ticho velkých okamžiků minulého století. Teď ale musí kamenný kůr nést něco úplně nového, a sice 33 tun svatovítských varhan. Kvůli novému nástroji museli památkáři prostor upravit – vyztužit nosné konstrukce, ale také šetrně nainstalovat do zdiva síť optických kabelů. Moderní hudební nástroj je dnes totiž tak trochu i počítač.

Policisté obvinili devět lidí a dvě společnosti pro podezření z podvodu při veřejných zakázkách a podvodu s dotacemi. Všichni jsou stíháni na svobodě. Kvůli zásahu skončil jako jednatel Mach.

Az előfizetők (de csak a Belső kör és Közösség csomagok tulajdonosai!) már szombat hajnalban hozzájutnak legfrissebb epizódunk teljes verziójához. A hétfőn publikált, ingyen meghallgatható verzió tíz perccel rövidebb. Itt írtunk arról, hogy tudod meghallgatni a teljes adást. Radikális fordulat dezodorügyben. NBA-döntő vs. vébé. Ellen-Trumpok halálfej-tetoválással. Tajtékpipa és gravity bong. Németh Balázs elmeállapota. Mikor csuknak már le valakit? Az osztrákoknál az alma is jobb. 00:00 Tartalomjegyzék. A 96 órás dezodor. Néhány támogatói jegy még van a mosdatlan Borízű Live-ra. 04:07 Személyiségi jogi védelem a Budaörsi uszodában. Nincs is semmi baj a dezodorokkal.08:52 Mérsékelt vébéhőemelkedés. A jó 94-s amerikai vébé. Diana Ross tizenegyese. Orbán Viktor megnézi az Üzbegisztán-Kolumbiát. 14:56 NBA-világbajnokság-együttállás. Wemby olvasgat. Spike Lee: Knicks in six. A csapatépítő Gregg Popovich. 21:19 A kosár, amivel a negyedik meccset megfordította a Knicks. Halt time show a Wu-Tang Clannel. Shaq bulizik a stúdióban. Amerikai sportrajongók politikai hovatartozása. 26:29 Platner, Talarico, Newsom és a demokraták laboratóriumban megalkotott ellen-Trumpjai. Mamdani Arsenal-drukker imája. Mamdani végigtippeli a vébét. 31:39 Kvíz: tajtékpipa. Te miből szívtál cracket? Pipázás a rendszerváltás körül. 36:33 Első 444 Bongépítő Verseny. Vödrözni egészséges! Bottle bag és egyéb gravity bongok. Cindy Breakspeare és Pascaline Bongo. 41:40 A Tisza-négyötöd robotosai. Ha jól megy a gazdaság, nem csukják le Orbánt. 43:04 Uj Péter Ausztriában. A Billában bezzeg rendes alma van! Cosmic crisp és Jack Herer. Az almabong óvodás szint. See omnystudio.com/listener for privacy information.

Flera personer visste att 21-årige James Stoneham ville mörda sin ex-flickvän, 20-åriga Adriana Donato. Ingen sa något till Adriana. Och därför gick hon med på att sätta sig i James bil den 23 augusti 2012. Kvällen som skulle bli hennes sista i livet.Manus av Sofie Karlsson. Klippning av Josefine Molén.Reklam. Om du gillar Mördarpodden kan du vara med och sponsra den på Patreon. https://www.patreon.com/user?u=10466265.Som tack får du tillgång till förhandlyssning och alla avsnitt från Richard Chase del 1 och framåt utan reklam. Vill du höra ett specifikt fall i podden? Önska dina fall i det här formuläret: https://docs.google.com/forms/d/e/1FAIpQLSfDlQxf9SgZyeGS-qFPaB4BP-L59lQhs7BbZACfwk7xSs-AFw/viewform?fbclid=IwAR0astYAY_SJLcst89FwKaPIeHHV9zlfAxEz6Cmrh37bbMwvMHGc8z5cwg4Det här är en podcast av Dan Hörning och Josefine Molén.Instagram: @mordarpoddenE-post: zimwaypodcast@gmail.comFölj Josefine Molén här:https://www.instagram.com/j.molenFölj Dan Hörning här:X: @danhorningInstagram: https://www.instagram.com/dan_horning/?hl=enYoutube: https://www.youtube.com/channel/UCV2Qb7SmL9mejE5RCv1chwgErik SegerstedtSpotify:https://open.spotify.com/artist/63q3l3pKBpvqEjUM5Vf1TG?si=fYtdOwIvTn6noQJW6ffPwwInstagram: https://instagram.com/e Hosted on Acast. See acast.com/privacy for more information.

Zemřela Zdena Mašínová, bylo jí 92 let. Na sociálních sítích to ve středu uvedl historik Petr Blažek či Konfederace politických vězňů. Mašínová byla dcerou generála Josefa Mašína, který za protektorátu bojoval v domácím odboji, a sestrou členů odbojové skupiny bratří Josefa a Ctirada Mašínových. Kvůli činnosti svých bratrů, kteří za dramatických okolností utekli na počátku 50. let z Československa na Západ, byla perzekvována komunisty.

Když Češi hráli na světovém šampionátu naposledy, mohli fanoušci sledovat zápasy pohodlně večer po práci – hrálo se totiž v Německu. Kvůli letošnímu mistrovství světa v Severní Americe si ale na dva ze tří zápasů základní skupiny budeme muset přivstat nebo vydržet vzhůru až do rána. Náročný program ale nečeká jen diváky, ale i samotné fotbalisty.

Chcete-li podpořit Studio Svobodného přístavu, můžete tak učinit v krypto i korunách! Pravidelná podpora a LN: https://opristavu.urza.cz/ BTC: bc1qwy8l3w0v826amd69h4awpt9hee6srxn4gk2cpg LTC: ltc1q2w2zezyj4anh3v428msf9kqvzelt76n62ys93h Číslo účtu: 2201359764/2010; variabilní symbol: 6 -------- Karolína se mi ozvala s přáním probrat potenciální regulace AI; nebo to půjde bez nich? Je AI skutečně hrozbou? Nebo naopak funkčním pomocníkem? Případně obojí? A co s tím? Pomohou tentokrát státy? Nebo spíš uškodí jako obvykle? Ve Studiu Svobodného přístavu se budeme věnovat AI z hlediska svobody; velká témata (zejména hrozby) bývají využívána k centralizaci, což samozřejmě zajímá i Karolínu balancující na hranici etatismu a anarchismu (čím dál blíž k tomu druhému). – Karolína Kváš (https://karolinakvas.cz/); tarotová rebelka; spisovatelka; cestovatelka; antropoložka; bloggerka; tvůrkyně; koučka – Urza (www.urza.cz); autor knihy Anarchokapitalismus; tvůrce Svobodného přístavu; spoluzakladatel a hlava Institutu Ludwiga von Misese; člen předsednictva Svobody učení; učitel ve svobodné škole Ježek bez klece

Äntligen är det dags för årets roligaste dopp och Sveriges blötaste folkrörelse: Kvällsdoppet! Lyssna på alla avsnitt i Sveriges Radios app. Ett nyfiket och underhållande aktualitetsprogram med lyssnaren i fokus.Sarit är i Studio 43 och tar badtempen på lyssnare runt om i Sverige. Samtidigt är Rasmus och Christer ute på Göteborgs vägar i varsin badbil för att plocka upp några badsugna och glada deltagare. Slutdestinationen är Delsjön, där det elfte årliga Kvällsdoppet går av stapeln!Dessutom har vi med oss lyssnaren Åse som försöker hinna fram till badplatsen Gubben i tid och Peter som meddelar att han kommer doppa sig i ett 14 grader kallt Vänern.

In this episode of Alexa's Input (AI), I sat down with Rob Shaw from Red Hat to talk about how AI inference evolved from a simple model serving problem into a large-scale distributed systems problem.We explored the infrastructure shifts behind modern LLM serving, including how vLLM and PagedAttention changed the economics and efficiency of inference, why KV cache management became one of the most important bottlenecks in production AI systems, and how orchestration layers like llm-d are emerging to coordinate distributed inference.We also discuss:how LLM inference differs from traditional model serving runtimesKV cache, prefix caching, and cache-aware routingwhy throughput and latency became major infrastructure challengeslong-context agents and repeated inference callsdistributed inference on Kubernetesintelligent routing, flow control, and load balancingprefill/decode disaggregationenterprise AI deployment realitiesvLLM has become one of the most important open-source projects in AI infrastructure, and llm-d represents a newer shift toward treating inference as a coordinated distributed system rather than just a single runtime problem.If you want to better understand the systems layer beneath modern AI applications, this episode is a deep dive into where inference infrastructure is heading next.General Podcast LinksWatch: ⁠⁠⁠⁠⁠⁠https://www.youtube.com/@alexa_griffith⁠⁠⁠⁠⁠⁠Read: ⁠⁠⁠⁠⁠⁠⁠⁠https://alexasinput.substack.com/⁠⁠⁠⁠⁠⁠⁠⁠Listen:⁠⁠ ⁠⁠https://creators.spotify.com/pod/profile/alexagriffith/⁠⁠⁠⁠More: ⁠⁠⁠⁠⁠⁠https://linktr.ee/alexagriffith⁠⁠⁠⁠⁠⁠Learn more about the host atWebsite: ⁠⁠⁠⁠⁠⁠https://alexagriffith.com/⁠⁠⁠⁠⁠⁠LinkedIn: ⁠⁠⁠⁠⁠⁠https://www.linkedin.com/in/alexa-griffith/⁠⁠⁠⁠⁠⁠Find out more about the guest at:LinkedIn: https://www.linkedin.com/in/robert-shaw-1a01399a/ Red Hat Articles: https://developers.redhat.com/author/robert-shawGithub: https://github.com/robertgshaw2-redhat ResourcesvLLM Website: https://vllm.ai/vLLM GitHub Repository: https://github.com/vllm-project/vllmllm-d Website: https://llm-d.ai/llm-d GitHub Repository - https://github.com/llm-d/llm-d KeywordsAI inference, VLLM, LMD, distributed inference, GPU optimization, open source AI, Kubernetes, multi-cluster deployment, AI infrastructure, enterprise AI AI infrastructure, Kubernetes, model optimization, speculative decoding, mixture of experts, AI deployment, performance tuning, AI systems, neural network scaling Key TopicsEvolution of vLLM and llm-dDistributed inference and routingGPU utilization and performance optimizationOpen source AI infrastructureEnterprise deployment challenges and solutions Standardization in Kubernetes for NIC exposurePerformance optimizations: quantization and speculative decodingMixture of experts architecture and parallelism strategiesFlow control and request scheduling in AI systemsEmerging hardware for AI inference, Cerebras processorReinforcement learning and AI system supportModular architecture of vLLM and ecosystem projects

En kväll i september ringer en tonårskille på dörren till ett radhus i Limhamn. Killen har med sig en Rambo-kniv och uppdraget att mörda irankännaren och regimkritikern Arvin Khoshnood. Lyssna på alla avsnitt i Sveriges Radios app. – Jag hade gjort mig beredd för mitt livs viktigaste kamp, jag var väldigt rädd men jag försökte hålla mig lugn och fokuserad för familjens skull, berättar han. Kvällen den 2 september 2025 är som vilken vanlig kväll som helst men den ska komma att förändra livet totalt för irankännaren och regimkritikern Arvin Khoshnood och hans familj. De ska snart gå och lägga sig när det ringer på dörren. Khoshnoods fru öppnar, utanför står en tonårskille som hon inte känner igen och säger att han söker Arvin. Hon stänger dörren och ropar på sin man, som direkt anar oråd. Vad ingen av dem vet just då är att tonåringen har med sig en kniv och uppdraget att hugga ihjäl Khoshnood. – Vi lever på hemlig ort, långt borta från vår familj och våra vänner och allting som vi hade byggt upp, säger Arvin Khoshnood idag.Nyligen dömdes tonåringen och flera andra unga killar för förberedelse till mord. Motivet är inte klarlagt men en av polisens teorier är att ”främmande makt” kan ligga bakom. – En efter en har vi avfärdat de olika scenarier som fanns, förutom ett enda scenario, att det här var ett morduppdrag beställt av den Islamiska regimen i Iran, säger Arvin Khoshnood. I avsnittet hör du även Moa, utredare vid grova brott hos polisen i Malmö och Gabriel Wernstedt, pressekreterare på Säkerhetspolisen. Programledare: Jenny Hellström och Linus LindahlLjud: Fredrik NilssonProducent: Jenny HellströmKontakt: p3krim@sverigesradio.seTipstelefon: 0734-61 29 15 (samma på Signal)

På med badmössan vi dyker vidare bland allt spännande, oväntat och underbart som har hänt apropå vatten! Lyssna på alla avsnitt i Sveriges Radios app. Ett nyfiket och underhållande aktualitetsprogram med lyssnaren i fokus.Vi hör bland andra Bosse som hade ett magiskt möte med en val under stjärnhimlen på Bermuda, och Pär som hittade ett bilvrak på sjöns botten. Dessutom delar Birgitta med sig av den dramatiska dagen då både hon och hunden Elvis gick genom isen!I extramaterialet firar vi två bröllopspar, laddar inför Kvällsdoppet med badbilspassageraren Joakim och tackar för den här vårsäsongen – innan vi säger hej till sommaren!

Pokud se vám přednáška líbila a shledáváte ji hodnotnou, prosím, pošlete dobrovolný příspěvek v krypto či korunách! Pravidelná podpora a LN: https://opristavu.urza.cz/ BTC: bc1qqs0sutxykl5h97xa0fcu4qaptvkvq5ecm52qn9 LTC: ltc1qpcnumcpvx2a77p0shkxwawc555nepy25l0n090 Číslo účtu: 2201359764/2010; variabilní symbol: 5 -- Květnová přednáška z cyklu Anarchokapitalismus bude vlastně spíš dialog dvou lidí než přednáška. Z „Konference Svobodného přístavu 2026: Svoboda včera, dnes a zítra“ možná znáte Matěje Krejčiříka, jenž představil kolektivistický anarchismus; to se stalo v rámci jednoho (zatím nejspíš neúspěšnějšího) pokusu o propojení české individualistické a kolektivistické anarchistické scény. Tuto myšlenku v sobě nosím už dlouho, ale pokaždé, když jsem se ji pokusil realizovat, selhala; kolektivu z AC254 se to daří výrazně lépe, byť naše pokusy o hledání společné řeči stojí obě strany mnoho sil, přesto však pokračujeme. Rádi bychom část našeho dialogu vedli veřejně; snažíme se totiž propátrat rozdíly mezi nazíráním na svět, čemuž můžeme zajisté také pomoci – a to jak svými otázkami, tak i snahou o pochopení a porozumění. Prezentace: https://prednasky.urza.cz/ll/ – Urza (www.urza.cz); autor knihy Anarchokapitalismus; tvůrce Svobodného přístavu; spoluzakladatel a hlava Institutu Ludwiga von Misese; člen předsednictva Svobody učení; učitel ve svobodné škole Ježek bez klece

VYPLŇTE NÁŠ PODCASTOVÝ PRIESKUM: http://www.zabavavpodcastoch.sk/prieskum Vláda už dlhšie avizuje zoznam prorastových opatrení, ktoré by mali pomôcť naštartovať slovenskú ekonomiku. Stále ich však oficiálne nezverejnila. Známe je len to, že ich má byť 54 a vláda ich má predstaviť na budúci týždeň aj v paragrafovom znení. Prvých osem opatrení rieši pomoc firmám s cenami energií, ďalších päť upravuje trh práce, desať zasiahne do oblasti daňovo-odvodovej politiky a tridsaťjeden chce zlepšiť podnikateľské prostredie. Podľa prezidentky KOZ SR Moniky Uhlerovej sú to zatiaľ len kozmetické úpravy bez výraznejšieho potenciálu naštartovať ekonomiku. Okrem toho dodáva, že balík prorastových opatrení zaváňa Sulíkovským kilečkom. „Sú tam rôzne návrhy na dereguláciu BOZP, kde naozaj neviem, o čom sa bavíme.“ Podobne je to aj pri pôvodnom návrhu skrátiť obedovú prestávku na 15 minút. „Tento návrh už v tom novšom balíku nie je, ale mám z toho pocit, akoby od začiatku mal len vytvárať dymovú clonu, aby sme sa bavili o tomto nezmysle a nevšímali si iné, podstatné veci,“ hovorí šéfka odborárov. Vláda v prvom balíku prorastových opatrení zrejme nepredstaví žiadne očakávané zníženie daní a odvodov. „Je tam len zavedenie tzv. príjmového testu v súvislosti s platením poistného SZČO. Povinnosť im vznikne, ak ich príjem presiahne 10,5 násobok životného minima.“ Pre odborárov je zásadné nastavenie životného minima, čo má vláda aj v programovom vyhlásení vlády. „Podľa nás by malo byť previazané s ukazovateľmi práce, ako je napríklad minimálna mzda a zmeniť tiež treba valorizačný mechanizmus,“ približuje Uhlerová. Koncom júna si odborári budú voliť nové vedenie a Monika Uhlerová má vyzývateľa v podobe europoslanca za stranu Hlas Branislava Ondruša. Kvôli možnosti uchádzať sa o post prezidenta KOZ už vystúpil z Hlasu, ale chce si nechať mandát europoslanca. Podľa Uhlerovej je v poriadku, že má vyzývateľa. „Som zástanca súťaže ideí. Lenže Branislav Ondruš je politik, a to mi príde neštandardné,“ komentuje jeho kandidatúru. Ondruš vstúpil do odborov len nedávno a naďalej chce zostať aj aktívnym politikom v Európskom parlamente. Uhlerová dodáva, že odbory majú byť nezávislé od akejkoľvek moci, vlády, či politickej strany. Majú zastupovať len záujmy pracujúcich a majú byť aj zdravou ľavicovou kritikou vlády. „Nie som fanúšičkou politiky kabinetného vybavovačstva. Odborári nemajú čo chodiť šúchať kľučky na ministerstvá a niečo vybavovať,“ dodáva šéfka KOZ. Rozhovor moderuje Eva Mihočková. V rozhovore sa dozviete: prečo vláda mešká s prorastovými opatreniami, aké zmeny môžu priniesť, prečo sa balík podobá na Sulíkove kilečko, čo znamenajú pribúdajúce hromadné prepúšťania, ako Uhlerová reaguje na vyzývateľa Ondruša. See omnystudio.com/listener for privacy information.

DeepInfra (https://deepinfra.com/) is serving over 5 trillion tokens a week and just closed a major raise backed by NVIDIA. In this episode of Pale Blue Nexus, host Yohann Calpu sits down with co-founder Nikola Borisov to unpack how DeepInfra is challenging the hyperscalers on price, why he compares inference optimization to Formula 1, and what the future of open source AI looks like.Nikola shares the technical playbook behind DeepInfra's aggressive pricing (including the famous Mixtral moment), how quantization and KV caching drive efficiency, the company's deepening partnership with NVIDIA on the Dynamo project, and why he believes the real demand for AI infrastructure is just getting started.We also get into the harder questions: prompt injection risks, the NemoClaw security model, hardware depreciation cycles, and what's actually overrated in today's AI infrastructure boom.Nikola bet early on open source inference. What's your bet? The Aloomii Playbook is the operator's manual for putting AI into relationship-driven businesses — not theory, not hype, just the system we use to run our own agent fleet (OpenClaw) and our clients'.Three editions for where you actually sit:Founder. Solopreneur. Operator Leader.→ Get the Playbook on Gumroad — $179 https://www.aloomii.com/playbook/ https://aloomii.gumroad.com/The Last 20% is the LinkedIn newsletter for operators who'd rather build the system than read about it, field notes, frameworks, and behind-the-scenes from running Aloomii's agent fleet (OpenClaw) and our clients' AI transformations. No theory. No hype. Just what's actually working. →Subscribe free on LinkedIn   / the-last-20-7445126674708451328  

I naturen kan man både njuta och förundras. Vi ger oss ut bland buskar, berg och stränder och hör berättelser om oväntade upplevelser ute i det fria! Lyssna på alla avsnitt i Sveriges Radios app. Ett nyfiket och underhållande aktualitetsprogram med lyssnaren i fokus.Karin vandrade rakt in i en nakenfotografering, Datten hittade en pyntad julgran mitt i skogen under svampplockningen och Fredrik berättar om en intensiv, dramatisk och blöt vildsvinsjakt!I extramaterialet pratar vi om hur både klurigt och roligt det är när Göteborgsvarvet tar över stan, om pappersbiljetter versus e‑biljetter och om peppen inför Kvällsdoppet!

Kattpsykologen Susanne Hellman Holmström är tillbaka i Kvällspasset för att svara på lyssnarfrågor om allt som har att göra med katter och deras beteenden! Lyssna på alla avsnitt i Sveriges Radios app. Ett nyfiket och underhållande aktualitetsprogram med lyssnaren i fokus.Hanna undra när en sibirisk katt blir fullvuxen, Tore vill ha tips på hur han kan få kattgrannen på sommarstället att sluta bråka med hans katt. Vi följer också upp hur det gått för lyssnaren Hussein som har använt Susannes tips för att få sin katt mer tillgiven!

מה הופך בקשה אחת ל-LLM למורכבת כל כך מאחורי הקלעים? איך מאות מיליארדי פרמטרים נדחסים על עשרות GPUs, ואיך כל ה-cluster הזה משרת אלפי משתמשים במקביל בלי להתפוצץ?אירחתי את מייק ארליכסון, אושייה בעולם הAI, ופירקנו את עולם ה-inference מבפנים: KV cache, batching, ההבדל בין prefill ל-decode, חלוקה של מודל בין GPUs שונים, ו-Mixture of Experts. דיברנו גם על למה זה הפך לאחד התחומים הכי קריטיים בעולם ה-AI - וגם איך נכנסים אליו אם אתם מהנדסים שרוצים להתחיל להריץ מודלים בעצמכם.האזנה נעימה, עמית בן דור.

Měla to být pompézní oslava 70. výročí soutěže, ve které se členové Evropské vysílací unie (EBU) předhánějí v boji o nejpopulárnější song roku. Místo toho Eurovize prochází možná nejtěžším obdobím ve své historii. Kvůli účasti Izraele čelí protestům i bojkotu tradičních účastníků. Jak je to s její proklamovanou apolitičností? Měří všem stejným metrem? A má vůbec budoucnost?Host: Tereza Povolná - redaktorka Seznam Zpráv Článek a další informace najdete na webu Seznam ZprávySledujte nás na sociálních sítích X, Instagram, Threads nebo Bluesky. Náměty a připomínky nám můžete psát na e-mail zaminutusest@sz.czHlasujte pro náš podcast v anketě Podcast roku.

Měla to být pompézní oslava 70. výročí soutěže, ve které se členové Evropské vysílací unie (EBU) předhánějí v boji o nejpopulárnější song roku. Místo toho Eurovize prochází možná nejtěžším obdobím ve své historii. Kvůli účasti Izraele čelí protestům i bojkotu tradičních účastníků. Jak je to s její proklamovanou apolitičností? Měří všem stejným metrem? A má vůbec budoucnost? Host: Tereza Povolná - redaktorka Seznam Zpráv Článek a další informace najdete na webu Seznam Zprávy Sledujte nás na sociálních sítích X, Instagram, Threads nebo Bluesky. Náměty a připomínky nám můžete psát na e-mail zaminutusest@sz.cz Hlasujte pro náš podcast v anketě Podcast roku.

Join us for a roundup from the IEEE PES T&D 2026 conference in Chicago, featuring interviews with exhibitors about the rise of 765 kV transmission, the looming matting shortage for large build-outs and the growing power demand from data centers. Vendors and technology providers discuss solutions for improving grid reliability and resilience. Hear conversations with Hubbell, Shemar, Sterling Site Access Solutions, Smart Power and Ayr Energy about local manufacturing, automated mat production, power-flow technologies and upcoming projects representing gigawatts of new capacity. For more coverage and videos, visit www.tdworld.com and tune into our sister podcast, T&D World Live. You can also listen to our latest episode in our ICYMI series, "Powering Reliability: Inside the 2026 IEEE PES T&D Show" on Podbean or in your favorite podcasting app. Thanks for listening!

In this episode, Philip Kiely, head of AI education at Baseten, joins us to unpack the fast-evolving discipline of inference engineering. We explore why inference has become the stickiest and most critical workload in AI, how it blends GPU programming, applied research, and large-scale distributed systems, and where the line sits between inference and model serving. Philip shares how research-to-production can move in hours, not months, and why understanding “the knobs” of inference—batching, quantization, speculation, and KV cache reuse—lets teams design better products and SLAs. We trace the inference maturity journey from closed APIs to dedicated deployments and in-house platforms, discuss GPU lifecycles, and survey today's runtime landscape, including vLLM, SGLang, and TensorRT LLM. Finally, we look ahead to agents and multimodality, making the case for specialized, workload-specific runtimes when performance and efficiency matter most. The complete show notes for this episode can be found at https://twimlai.com/go/766.