Podcasts about Gram

Game server hosting is the layer nobody thinks about until it breaks. In this episode, Kalie Moore sits down with Max Podkidkin, Co-Founder and CEO of BisectHosting, to unpack how a Minecraft server that went dark for three days turned into a 15-year infrastructure business serving hundreds of thousands of customers across more than 140 countries. Max put the first hardware on a credit card, recruited a co-founder he met in his own Minecraft community, and built the company to nearly 100 employees without ever raising outside capital. They get into what those early years actually looked like, including hiding a laptop on a cart at a factory job to answer support tickets, the two years he and Andrew worked together over text before meeting face to face, and how a US and UK split gave them 24/7 coverage with a team of two.The conversation also covers how the hosting business really works, from the economics of renting out resources to why running a server at home costs more than most people think. Max explains why BisectHosting has turned down repeated acquisition offers while private equity consolidates the market around them, how the team gets into a studio's Discord and pre-tests server files before working with a game studio, and what 15 years of hosting communities has taught him about why modding and UGC keep games alive long after launch. He also shares the studio's approach to AI, including Biko, the assistant built into their server panel that has cut support tickets by double digits without a single person losing their job.Big thanks to Heroic Labs for making this episode possible. Gram Games used Heroic Labs to tailor offers and events by player cohort, calling it "critical" in a post-IDFA world where retention is everything. Read how Gram merged personalization and advanced social features into a core system of their live operations: https://heroiclabs.com/blog/gram-games-case-study/?_gl=1*709y52*_up*MQ..*_ga*MTI0MTAzNTM2Ny4xNzg1MjI4NzQ2*_ga_R9WLGSZ1KN*czE3ODUyMjg3NDUkbzEkZzAkdDE3ODUyMjg3NDUkajYwJGwwJGgw We'd also like to thank Overwolf for making this episode possible! Whether you're a gamer, creator, or game studio, Overwolf is the ultimate destination for integrating UGC in games! You can check out all Overwolf has to offer at https://www.overwolf.com/.If you like the episode, please help others find us by leaving a 5-star rating or review! And if you have any comments, requests, or feedback shoot us a note at podcast@naavik.co.Who's On:Guest - Max Podkidkin: https://www.linkedin.com/in/max-podkidkin/Host - Kalie Moore: https://www.linkedin.com/in/kaliemoore/ Watch the episode: YouTube ChannelFor more episodes and details: Podcast WebsiteFree newsletter: Naavik DigestFollow us: Twitter | LinkedIn | WebsiteSound design by Gavin Mc CabeLinks mentioned: https://www.bisecthosting.com/

It's all questions where all the answers are numerical! This episode's topic: THE NUMBERS GAME Sponsored by LearnClash, the quiz duel app that helps you remember what you learn: learnclash.com/partners/budds Fact of the Day: A potential side effect of Penicillin and related antibiotics is a harmless rash that leads many to a false belief that they are allergic. Triple Connections: Book, Gram, Tube THE FIRST TRIVIA QUESTION STARTS AT 01:47 SUPPORT THE SHOW MONTHLY, LISTEN AD-FREE FOR JUST $3 A MONTH: www.Patreon.com/TriviaWithBudds INSTANT DOWNLOAD DIGITAL TRIVIA GAMES ON ETSY, GRAB ONE NOW! GET A CUSTOM EPISODE FOR YOUR LOVED ONES: Email ryanbudds@gmail.com Theme song by www.soundcloud.com/Frawsty Bed Music: "Laser Groove" Kevin MacLeod (incompetech.com) Licensed under Creative Commons: By Attribution 4.0 License http://creativecommons.org/licenses/by/4.0/ http://TriviaWithBudds.comhttp://Facebook.com/TriviaWithBudds http://Instagram.com/ryanbudds Book a party, corporate event, or fundraiser anytime by emailing ryanbudds@gmail.com or use the contact form here: https://www.triviawithbudds.com/contact SPECIAL THANKS TO ALL MY AMAZING PATREON SUBSCRIBERS, INCLUDING: Samantha Wheeler Boomer Cates Mark Kloppenburg Cadi Snow-Brine Amber Shiels Alan Kreisel Rich Sommer Joe Heiman Waqas Ali Logan Booker Bringeka Sam Nathan Stenstrom Brooks Martin Robyn Price Gee Brian Clough Charles Glanville IV Lauren Schuette Evan Lemons AnneMarie Mattacchione Yves Bouyssounouse Kenny Zail York yates Gay Geek Fabulous Mollie Dominic Nathalie Avelar Natasha raina leslie gerhardt Diane White Youngblood Trophy Husband Trivia Lynnette Keel Lillian Campbell Jerry Loven Jamie Greig Gail Lancman Jeremy Yoder Adam Jacoby rondell Adam Suzan Tiffany Poplin Bill Bavar Sarah Daniel Hoisington Keith Martin Sue First Steve Hoeker Jessica Allen Lauren Glassman Brian Williams Brett Livaudais Linda Elswick Carter A. Fourqurean Justly Maya Brandon Lavin Kathy McHale Chuck Nealen Courtney French Nikki Long Mark Zarate Laura Palmer  JT Dean Bratton Kristy Erin Burgess Trenton Sullivan Jen and Nic Michael Redman Timothy Heavner Jeff Foust Richard Lefdal Myles Bagby Jenna Leatherman Vernon Heagy Albert Thomas Kimberly Brown Tracy Oldaker Sara Zimmerman Madeleine Garvey Jenni Yetter Patrick Leahy Dillon Enderby James Brown Christy Shipley Clayton Polizzi Alexander Calder Ricky Carney Paul McLaughlin Willy Powell Robert Casey Matthew Frost Brian Salyer Greg Bristow Megan Donnelly Jim Fields Mo Martinez Luke Mckay Simon Time Feana Nevel Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Gram-negative resistance is beautifully intricate, scientifically elegant, and very complex. Nothing is more exciting than the recently approved and bursting pipeline of beta-lactamase inhibitors. Listen to Drs. Sam Aitken and Christophe Le Terrier, along with host Dr. Erin McCreary, describe the nuanced enzymology and other antibacterial properties of beta-lactamase inhibitors. This episode also discusses the concepts of stand-alone inhibitors, resistance development, and more. If you want to learn more about sulbactam, clavulanic acid, tazobactam, avibactam, relebactam, vaborbactam, durlobactam, enmetazobactam, zidebactam, nacubactam, ledaborbactam, taniborbactam, and xeruborbactam – this episode is for you! Funding for this podcast was provided by Shionogi Inc. References: 1.      Le Terrier C, Nordmann P, Buchs C, et al. Wide dissemination of Gram-negative bacteria producing the taniborbactam-resistant NDM-9 variant: a One Health concern. J Antimicrob Chemother. 2023;78(9):2382-2384. doi:10.1093/jac/dkad210 2.      Le Terrier C, Nordmann P, Buchs C, Poirel L. Effect of modification of penicillin-binding protein 3 on susceptibility to ceftazidime-avibactam, imipenem-relebactam, meropenem-vaborbactam, aztreonam-avibactam, cefepime-taniborbactam, and cefiderocol of Escherichia coli strains producing broad-spectrum β-lactamases. Antimicrob Agents Chemother. 2024;68(4):e0154823. doi:10.1128/AAC.01548-23 3.      Le Terrier C, Freire S, Viguier C, Findlay J, Nordmann P, Poirel L. Relative inhibitory activities of the broad-spectrum β-lactamase inhibitor xeruborbactam in comparison with taniborbactam against metallo-β-lactamases produced in Escherichia coli and Pseudomonas aeruginosa. Antimicrob Agents Chemother. 2024;68(6):e0157023. doi:10.1128/AAC.01570-23 4.      Le Terrier C, Mlynarcik P, Sadek M, Nordmann P, Poirel L. Relative inhibitory activities of newly developed diazabicyclooctanes, boronic acid derivatives, and penicillin-based sulfone β-lactamase inhibitors against broad-spectrum AmpC β-lactamases. Antimicrob Agents Chemother. 2024;68(11):e0077524. doi:10.1128/AAC.00775-24 Learn more about the Society of Infectious Diseases Pharmacists: https://sidp.org/About Instagram: @SIDPharm ( https://www.instagram.com/sidpharm/) or @breakpointspodcast_sidp (https://www.instagram.com/breakpointspodcast_sidp/) https://www.instagram.com/breakpointspodcast_sidp/?hl=en Facebook: https://www.facebook.com/sidprx LinkedIn:  https://www.linkedin.com/company/sidp/ SIDP welcomes pharmacists and non-pharmacist members with an interest in infectious diseases, learn how to join here: https://sidp.org/Become-a-Member Listen to Breakpoints on iTunes, Overcast, Spotify, Listen Notes, Player FM, Pocket Casts, Stitcher, Google Play, TuneIn, Blubrry, RadioPublic, or by using our RSS feed: https://sidp.pinecast.co/https://sidp.pinecast.co/

En este episodio de gramática y lengua española vamos a practicar con los diminutivos, ya sabes, estos sufijos que añadimos a algunas palabras para hablar de su tamaño, para mostrar afecto y, aunque suene sorprendente, desprecio.  

Author brand strategist Jessica Sorentino joins the show to talk about the overlooked second career every writer faces: building a brand and platform after the book is done. She and Brittani dig into the difference between brand and marketing, why in-person connection drives book sales more than social media, how to use AI without losing your voice, and when it's actually okay to call yourself a writer. Resources: The Meeting Place Membership Rock The Reels 1:1 Coaching Free Client Welcome Guide Additional Trainings and Resources Connect with Brittni: Follow me on the Gram - @brittni.schroeder Join my Facebook Group  Visit my website Subscribe to my Youtube You can find the complete show notes here: https://brittnischroeder.com/podcast/author-brand-strategist-with-jessica-sorentino

We chat writing a song in pieces, being sober(ish), having a niche genre bias, song writing without the struggle, talking about old shit, and more!   You can help support the show on our PATREON for as little as $1 a month! Double down to bump it to $2 a month and you'll get an extra episode every week!   Join the fun on our Facebook group!   Follow us on the fuckin' Gram!   Subscribe to our YouTube Channel for the video version of the show, demos, vlogs, and more!   We have shirts available at The Jerk Store!   Check out our band Plane Without a Pilot   Hosted by Brian Gower and Kyle McIntyre

Det er ikke størrelsen det kommer an på sies det, men at en muskel på 7 gram skulle bli en viktig del av et stort vendepunkt i livet mitt, det var jeg rett og slett ikke forberedt på... Hosted on Acast. See acast.com/privacy for more information.

For years, business owners have been laser-focused on SEO — but the way people search is changing fast. In this episode, we break down Answer Engine Optimization (AEO), the strategy every business owner needs to understand as more people turn to AI tools like ChatGPT and Claude to find answers instead of scrolling through search results. We cover what AEO actually is, how it works alongside traditional SEO, and seven practical ways to start optimizing your content so AI tools recognize your business as a trustworthy source. Whether you're a solopreneur or scaling a team, this episode will help you future-proof your marketing strategy for the AI era. Resources: The Meeting Place Membership Rock The Reels 1:1 Coaching Free Client Welcome Guide Additional Trainings and Resources Connect with Brittni: Follow me on the Gram - @brittni.schroeder Join my Facebook Group  Visit my website Subscribe to my Youtube You can find the complete show notes here: https://brittnischroeder.com/podcast/what-is-aeo-the-new-search-strategy-every-business-owner-needs-to-know  

It's 6am and we're giving you a REAL-TIME update on where the podcast, where our their lives are headed, and everything you didn't need to know but now you know lol. We're talking...the financial realities behind keeping the show going, our weekend highlights, why the podcast rebrand, and what Mac is naming her baby. This is just a chit chat that we hope you love! Thank you to today's sponsor! Hiya Health... our favorite vitamins + more to give our kiddos! Get 50% off your first order using our exclusive link! Pre-Order Our Book: This Wasn't The Plan Follow us on the Gram: @forthegirl___ Learn more about your ad choices. Visit megaphone.fm/adchoices

Matters Microbial #136: When Bacteria Open the Wrong Door August 11, 2026 Today Dr. Michael D.L. Johnson, Associate Professor of Immunobiology at the University of Arizona,  joins the quality quorum today to discuss copper, microbiology, and a possible new strategy to fight bacteria that cause disease. Host: Mark O. Martin Guest: Michael D. L. Johnson Subscribe: Apple Podcasts, Spotify Become a patron of Matters Microbial! Links for this episode A link to Hartiful, the Etsy artist who custom creates glow in the dark (and many other kinds) pins. A link to the essential science artist Beatrice the Biologist. A link to World Microbiome Day. A link to Micropia, the microbiology museum in Amsterdam.   A recent article on the growing threat of antibiotic resistance among disease causing bacteria. A video demonstrating the Kirby Bauer procedure, which shows antimicrobial effects and zones of inhibition. A link to an earlier #MattersMicrobial episode with Dr. Michael Schmidt describing copper and microbes. A review article describing the impact that copper can have on microbes.  A review article on Streptococcus pneumoniae, the organism with which Dr. Johnson works. An video celebrating the role that S. pneumoniae had in determining that DNA was the genetic material prior to World War 2. An overview of siderophores that help microbes obtain needed iron. An article about redox activity and copper and microbes. The relationship between metal toxicity and antimicrobial activity. The role that copper plays in agriculture. The differences between Gram positive and Gram negative bacteria. An article on efflux pumps and metal toxicity in microbes. An article describing how iron binding proteins also bind copper. An article describing how zinc interacts with microbes. A wonderful description of Dr. Johnson's "Trojan Horse" approach to battle disease causing microbes. A relevant review article by Dr. Johnson and his colleagues discussing copper and microbes. An overview of the "Hypothesis Fund," which supports unusual ideas in science. A video of Dr. Johnson describing his research. Dr. Johnson's faculty website. Dr. Johnson's research laboratory website. Intro music is by Reber Clark Send your questions and comments to mattersmicrobial@gmail.com

We had friends over for game night and my friend was repositioning the lime on her margarita to get the perfect photo. And she doesn't even do social media. So I said, "Do it for the joy." And that's when it hit me — what a great motto to live by.In this episode I'm talking about the moments that feel like the meaning of life (the farmer's market mornings, the perfect margarita by the fire), why we're so conditioned to capture and package everything for other people instead of just being in it, and the prompt that will help you come back to what actually matters: do it for the joy.Key Takeaways:The margarita moment that sparked this whole downloadWhy we're so conditioned to do things for the gram (or for other people) instead of for the joyThe farmer's market morning where I connected with a stranger over strawberriesWhat I want to remember when I look back on my life (not the big milestones — the rich, small moments)The note I keep in my phone of moments I want to bottle upThe prompt: do it for the joy, not for the gramScaling to multiple 6-figures? Apply for the CEO Mastermind here.Stabilizing your revenue before your scale? Get inside The Inner Circle here.Integrate AI systems & workflows into your business this weekend: Save $100 on Claude in a WeekendTry out the software I use to run my ENTIRE business: 30 day free trial for Go High Level

We go deep into the weeds about EQ, barefoot shoes, Gen Z terms, and more!   You can help support the show on our PATREON for as little as $1 a month! Double down to bump it to $2 a month and you'll get an extra episode every week!   Join the fun on our Facebook group!   Follow us on the fuckin' Gram!   Subscribe to our YouTube Channel for the video version of the show, demos, vlogs, and more!   We have shirts available at The Jerk Store!   Check out our band Plane Without a Pilot   Hosted by Brian Gower and Kyle McIntyre

Multidrug-resistant Gram-negative infections remain one of the most urgent public health challenges worldwide, with some infections approaching the limits of available treatment options. In this special edition of Editors in Conversation, we celebrate the remarkable 10-year tenure of Dr. Laurent Poirel, whose pioneering work has profoundly shaped our understanding of antimicrobial resistance. Through his unique perspective, we will explore the evolving landscape of resistance in Gram-negative organisms, reflect on the major advances and challenges of the past decade, and discuss what the future may hold for this rapidly changing field.  View this episode: https://youtu.be/3OuzU4HjZcY Topics discussed: The most important events in Gram-negative resistance in the last 10 years How the antibiotic pipeline has responded to the challenge of Gram-negative resistance How the field would move in the next 10 years. Guest: Laurent Poirel, Ph.D. Senior Researcher, Medicine Section. University of Fribourg, Fribourg, Switzerland, Editor AAC 2016-2026. Links: This episode is brought to you by the Antimicrobial Agents and Chemotherapy Journal.  Visit asm.org/aac to browse issues and/or submit a manuscript. If you plan to publish in AAC, ASM Members get up to 50% off publishing fees. Visit asm.org/joinasm to sign up.

En este episodio de gramática y lengua española vamos a ver algunas de las principales estructuras del modo subjuntivo para hablar de dudas, hipótesis, posibilidades… 

In this episode, bestselling author and financial coach Liz Carroll shares how shifting your money mindset can transform your financial future. She discusses why women often struggle with limiting beliefs around money, the importance of creating a spending plan instead of a budget, and how redefining wealth can lead to greater confidence and fulfillment. Liz also shares insights from her book Rich & Radiant, offering practical strategies to help women move from financial stress to lasting financial freedom.   Episode Freebie! Follow Liz on Instagram Liz's Website   Resources: The Meeting Place Membership Rock The Reels 1:1 Coaching Free Client Welcome Guide Additional Trainings and Resources Connect with Brittni: Follow me on the Gram - @brittni.schroeder Join my Facebook Group  Visit my website Subscribe to my Youtube You can find the complete show notes here: https://brittnischroeder.com/podcast/rich-radiant-with-liz-carroll

Watch the full episode on YouTube:We first covered Baseten last year when DeepSeek mania was at peak hype. Now they have raised a monster $13B round and become one of the new cohort of AI Infra decacorns that are (with Nvidia, Intel, and the semis complex) chief beneficiaries of the Inference Inflection. We return to Baseten at the peak of the 2026 edition of Open Weights debate. Ali has published a viral breakdown of Kimi K3:And since you last saw him, Philip has spoken at AI Engineer and written the definitive book on Inference Engineering spotted all over SF:Three years ago, inference engineering barely existed as a category.Today, it is one of the most critical disciplines in AI. Inference engineering inherently tackles a different question than standard model training: “How do you turn those weights from training into a product that is fast, reliable, and affordable at scale?” Focusing on these creates an entirely new optimization problem.In one recent GLM-5.2 experiment, quantizing more of the model actually preserved its benchmark quality while increasing throughput by 20%, because the errors introduced in different layers could cancel each other out.Inference is no longer just the final step after training. It is becoming its own engineering discipline, with its own research problems, infrastructure, and increasingly specialized roles.In this episode, Baseten's Philip Kiely and Ali Taha join swyx and Vibhu to explain what actually happens after a new open model is released and what it takes to turn “we generated a token” into a fast, reliable, production-ready API.We go deep on cache-aware routing, disaggregated prefill and decode, quantization, speculative decoding, KV-cache movement, model parallelism, GPU kernels, and the race to make frontier models up to 10× faster. Philip and Ali explain why inference optimizations can still produce gains of 20%, 100%, or even 200%; how quantization errors can cancel one another out; why identical weights can behave differently across clusters; and how Baseten grafted a Kimi vision encoder onto GLM-5.2 without changing the underlying language model.The conversation then expands beyond LLMs into NVIDIA Dynamo, mega kernels, Rubin, AI-specific chips, local inference, video generation, diffusion versus autoregressive models, and the enormous compute barrier to generating coherent long-form video. Finally, we explore the convergence of training and inference, continual learning through persistent KV cache, and the emerging loop where models help optimize the infrastructure that runs them.We discuss:* What happens when a 200,000-token request enters an inference system* Cache-aware routing and reusing previously computed KV cache* Why prefill and decode are increasingly handled by different GPUs* When dedicated deployments become cheaper and more reliable than shared APIs* How speculative decoding uses a smaller model to accelerate a larger one* Tool calling, structured outputs, and what LLMs actually do* What it takes to support a new open model on day zero* Grafting Kimi's vision encoder onto GLM-5.2* Retrofitting inefficient model layers with components from other architectures* Why models sometimes collapse into repeating the same token* How hardware, kernels, and race conditions create nondeterministic failures* Preserving model fidelity while making inference faster* How quantization errors can cancel each other out* Why inference optimizations still deliver gains of 20%, 100%, and 200%* How optimized serving can make a model up to 10× faster* NVIDIA Dynamo, KV-aware routing, and distributed model serving* Speculative decoding the speculative decoder* Why local AI is about making models less dumb while data-center AI is about making them less slow* Tensor, expert, and pipeline parallelism across GPUs* Hardware-aware model design, auto-tuning, and the case against mega kernels* Rubin and why inference is becoming a systems problem* Whether modern GPUs are evolving into programmable AI ASICs* Why enormous models like Kimi K3 require GB300-class hardware* Why open-source video generation still trails Veo, Kling, and other closed models* The quadratic attention bottleneck behind long-form AI video* Autoregressive video, real-time generation, and compounding quality drift* Why future video systems may combine autoregressive and diffusion architectures* Training for inference and inference for training* Continuous post-training, deployment, evaluation, and improvement loops* How GLM-5.2 helped optimize the kernels serving GLM-5.2 itself* Why faster networking could unlock dramatically faster decoding* Continual learning, KV-cache compaction, and persistent model memoryShow Notes* How to build a day-0 API for Kimi K3* 22580: From GPT2 to Kimi3, ExplainedPhilip Kiely* LinkedIn: https://www.linkedin.com/in/philipkiely* X: https://x.com/philipkiely* Inference Engineering: https://www.baseten.co/inference-engineering/Ali Taha* LinkedIn: https://www.linkedin.com/in/aliestaha/* X: https://x.com/waterloointernTimestamps00:00:00 Introduction and the 200K-Token Prompt00:03:18 Dedicated Deployments, Speculative Decoding, and Tool Calling00:11:26 Launching Production-Ready Open Models00:19:06 Model Retrofits, Failure Modes, and Nondeterminism00:28:22 Quantization and Canceling Errors00:32:15 The Race to 10× Faster Inference00:40:48 Dynamo, Speculation, and Local vs. Data-Center AI00:50:18 Model Parallelism, Auto-Tuning, and Mega Kernels01:00:55 Rubin, GPUs vs. ASICs, and Custom AI Chips01:10:03 Giant Models and the Limits of GPU Memory01:12:42 AI Video, Quadratic Attention, and Autoregressive Generation01:21:47 Audio, Images, and Diffusion Models01:27:32 Training, Self-Optimizing Models, and Continual Learning01:40:06 Closing ThoughtsTranscriptIntroduction: Baseten, Waterloo Intern, and Inference EngineeringSwyx [00:00:00]: Okay, we're here in the studio with Philip, old friend from Inference Engineering, the book, as well as Baseten and everything that you've done, you and I have done before, as well as Ali. Welcome.Ali [00:00:15]: Pleasure to meet you.Swyx [00:00:15]: Waterloo intern.Ali [00:00:16]: Waterloo intern, always.Swyx [00:00:17]: When did you get “Waterloo intern” as a handle?Ali [00:00:19]: As a handle? Oh.Ali [00:00:20]: I think the rebranding happened mid-March. When I saw it was open, I was like, “I have to take it. Up for grabs.”Philip [00:00:26]: The problem is that Ali is really good at his job and is not gonna be an intern much longer.Philip [00:00:30]: So we have to figure out who's gonna get the handle.Ali [00:00:33]: Well, I'll pass the torch over to the next intern.Swyx [00:00:34]: Oh, okay. It can be, like, you just pass it to another Waterloo grad.Ali [00:00:37]: To another Waterloo intern. No, bruh.Philip [00:00:39]: Yeah.Ali [00:00:39]: Intern.Swyx [00:00:40]: Intern, yeah.Ali [00:00:40]: And no.Philip [00:00:41]: You gotta get an intern from Waterloo.Ali [00:00:42]: Yeah, I've gotta get an intern from Waterloo.Swyx [00:00:44]: Right.Ali [00:00:44]: But they have to follow the path.Swyx [00:00:45]: Oh, it could, but it could come from Baseten, so it's like whoever Baseten gets from Waterloo.Ali [00:00:48]: Right.Swyx [00:00:49]: Has the title of Waterloo.Ali [00:00:50]: It stays in the ecosystem.Philip [00:00:51]: Exactly.Ali [00:00:52]: Halfway through the internship, you either get it or you're out.Philip [00:00:55]: You should also do, like, a big graduation ceremony where you change the handle.Ali [00:00:59]: Just say it.Philip [00:00:59]: For everybody.Swyx [00:01:00]: You guys are good at ceremonies, clearly. We had a nice launch of the book, very successful. But before we get into all that, I wanna start off with a fun question for you. Okay, you're an expert inference engineer. What happens when I send a long query, say two hundred thousand tokens into Baseten's inference? What's the process of query through GPU model routing, balancing, all that? What is all the stuff that we don't think about?Long Context Requests, KV Cache, and Cache-Aware RoutingPhilip [00:01:26]: With a long query specifically, the first thing that I'm gonna ask is, “Have you sent me this query before, or at least part of it?” and I really hope you have, because it's gonna be a lot easier for me and a lot cheaper for you. So the first thing that we're gonna look at is some cache-aware routing, where we're going to see, we probably have a number of instances, a number of replicas up serving whatever model you're hitting. We want to send this one to something with, number one, available prefill workers, and number two, ideally some cached input already there so that we can skip prefill on at least part of these two hundred thousand tokens. If you're doing two hundred thousand tokens, it's probably coding or a multi-turn agent or something where you would expect to have that cached. If you don't, we're gonna have to send it to a prefill worker. We've at least on certain models disaggregated prefill and decode, so you're going to have one set of GPUs that's solely going to process the input, create the KV cache, and get you your first token, and then that's going to be passed over to a separate set of GPUs, which is going to run decode. We're going to iteratively make those tokens. We're probably going to have some speculator model in front of that. I'm going to assume that you're doing coding, and because of that, our speculator model, which assumes you're doing coding, is gonna have a high draft token acceptance rate. If I'm wrong and you're asking me to summarize every Harry Potter book, it's gonna be slower. And then we stream that output to you and account for it, charge you, a couple of pennies and say, “Hey, would you like to send another one?”Swyx [00:03:04]: Except Baseten doesn't charge by pennies.Philip [00:03:07]: Well, yeah, we charge. I'm assuming that we're talking about the public model APIs. If you are setting up a dedicated deployment, then yeah, it's not pennies.Public APIs vs. Dedicated DeploymentsSwyx [00:03:18]: Yeah, one of the key differentiators when I was talking with Baseten initially was that people who want very high volume just need to rent by the box, ‘cause then it's up to you to figure out how to saturate the box.Ali [00:03:31]: And more often than not, it's, like, way cheaper if you're pushing, like, millions of tokens per hour, if you just pay per hour instead of pay per token.Philip [00:03:37]: Yeah, they do. I think that we've increasingly seen a lot of demand for the pay per token APIs, just because everyone wants to try open models, and then once they find a use case that's really sticky, then they move over to dedicated.Swyx [00:03:51]: Is there a best practice on when it's time to swap over?Philip [00:03:54]: Couple reasons. Yeah, reliability, that's a big one, right?Ali [00:03:57]: Like, if they have a very specific use case, they want you to train something specifically for them, like they want their own spec dec, for instance, for their own traffic.Swyx [00:04:04]: Spec dec is speculative decoding.Speculative Decoding and Custom SpeculatorsAli [00:04:05]: Speculative decoding, yeah.Swyx [00:04:07]: You have to explain.Ali [00:04:07]: Sorry. Like, speculative decoding is like, if you have a huge model, right? And so the model is going to be generating one token at a time every single turn, every single forward pass. So we attach, like, this little, like, parasite, like this layer that goes on top of the model, and this model just has to predict. It does three very fast autoregressive forward passes, and it will predict, like, three certain tokens, and then you do one forward stage over the entire original model in order to see if those predictions were correct or not, and then you accept them or you reject them. Now, this draft model is traffic specific, so if you, like, Philip said, if you're summarizing Harry Potter books, I can train exclusively that draft model on Harry Potter books, and I can guarantee you that I'm gonna accept the three tokens every single time. And so with that case, I increase your decode speed. I wouldn't be able to provide this to you if you're a shared endpointSwyx [00:04:53]: YeahAli [00:04:53]: ‘cause I have no idea if you're doing Harry Potter, if you're doing coding, if you're doing English. We don't know. Also, there was a thing in the book that mentioned that if they really cared about a specific threshold, chapter four, I think. Do you remember that?Philip [00:05:06]: Yeah. The things that you can do is you can set a specific, like, batch sizing, a specific, like, parallelism strategy if you're trying to optimize for, like, throughput versus latency. You can. Maybe a NVFP4 quant doesn't pass your benchmarks and you wanna run a model at higher precision, you could do that. There's just a bunch of reasons why you might wanna have your own endpoint and the biggest one, of course, just being, like, you don't have to deal with someone else doing a hundred million tokens of benchmarking traffic at the endpoint when you happen to be trying to serve your users.Swyx [00:05:40]: Yeah. I think one thing that is. That is a classic journey. Like, it's people is asking the, what happens when you type Google into the browser. Tool calling, is that just, you're generating JSON or is there more complication beyond that?Tool Calling, JSON, and Structured OutputsAli [00:05:58]: Certain customers that we have, they have their own post-trained models, and so they demand a tool calling that's not just, like parse a file or go find the weather. It's something that's very specific and you have to do post-training on this. And if the post-training on the model is not good or if the quantization after the post-training to get the inference to be fast, the model will struggle reading the JSON file and reading the tool calling. But it doesn't require its own like sandbox. It's not like it's going to use that tool calling to like escape a sandbox or like it doesn't have to be contained. It can just be a normal dedicated deployment. The challenge with tool calling more and more seems to be that the companies want certain tool calling which is a very sensitive thing to train. And because you're dealing with all of the JSON outputs, if it doesn't like close the end of the request in a very certain manner, you end up with a model that did the tool calling and like the thinking and so as a result of that, it didn't see the result and just hallucinated the result as it decoded. That seems to be the most challenging thing with tool calling, not really the sandboxes model.Philip [00:06:56]: Yeah, that's a challenge on the training side and then on the inference side, there's work that you can do to scope the possible output. So we published this at this point close to two years ago, the solution to this problem which is you make a state machine and you use that to constrain the output to a specific format. So this is the structured output problem. If you remember backSwyx [00:07:27]: Yeah, the specific grammar is,Philip [00:07:29]: Yeah, exactlySwyx [00:07:30]: GML had this thing.Philip [00:07:31]: Yeah. So it's like the old-school “make sure this is only JSON”, return only JSON orSwyx [00:07:38]: YeahPhilip [00:07:38]: Grandma's gonna die type of prompts.Swyx [00:07:39]: Is it BNF grammar? At some point OpenAI had released a thing that was like, yeah, if you want to constrain your output, write BNF grammar, back as NOR.Philip [00:07:47]: In our inference system, it's just a specified output format. And you get the guarantee that your output's gonna be structured along that format. And so applying that to tool calls can like help cut down on. You can still call the wrong tool or call no tool. It doesn't solve the certainty problem but it at least solves the output structuring problemSwyx [00:08:10]: YeahPhilip [00:08:10]: Within tool calls.Swyx [00:08:12]: And MCP is just another form of tool, right.Philip [00:08:14]: Yeah, exactly.Swyx [00:08:15]: As far as there's no special thing there.Philip [00:08:16]: The thing I'm always like explaining to people is the LLM is not capable of doing anything. It's only capable of making suggestions of what to do and then if those suggestions are formatted in a certain way and applied to a system that knows what to do with them, then an action occurs.Swyx [00:08:32]: Yeah. Part of the fun stuff is, this is solved outside of tool calling too. Like in an agent loop if the output is not correct or you're right, like reasoning, tool calling was done in the reasoning trace, just be like, “Oh, I don't know what to do. Let me just try again.” And it might get there after a few tries. And on your point of training, sometimes this is harder in smaller models, so you don't have the same exact quality outputAli [00:08:56]: Right.Swyx [00:08:57]: When you just swap from a big model, right?Ali [00:08:59]: Yeah. I will say that, before, I think we need to go back to inference engineering proper.Ali [00:09:04]: But, I had expected that something would replace JSON because it's hard to stream JSON ‘cause JSON must be complete and you must have open and close brackets and everything. So it's hard to parse something or validate something while it's being streamed. So people invented all sorts of things that are like, I forget the name of some of these alternatives, but it's something like TOML, something like YAML. But JSON seems to be dominant still.Philip [00:09:30]: The JSON outputs aren't that long, right? Like you could have a long-- ‘cause tool calls also contain the arguments in them and perhaps for a certain tool you might pass like a very long argument. But my impression of the median tool call is that it's a relatively small number of tokens, right? So I would expect that speculators are generally fairly good at something as formatted as JSON. And so you would have like a pretty fast decode step there and that the streaming wouldn't be as valuable, but maybe I'm wrong about that.Ali [00:10:02]: I think you're also bounded by the software or that the model is gonna integrate with if the software is built with JSON for the tool calls or if the company that you'- if your customer says that this is how our software works and our tools are interfaced with JSON, you can ask them to like, change their software and say like, “Yeah, this is gonna be better for the model.” but like with the right training shouldn't be that much of a difference. Also more profitable if it outputs more tokens probably.Swyx [00:10:25]: Depends on your business model.Swyx [00:10:27]: It really depends. But I will say that, as a writer with like experience a lot with generated output, I do try to move from text to JSON text which is very long JSON, right? Like there's paragraphs in every field because I'm trying to structure it, right?Philip [00:10:44]: Right.Swyx [00:10:44]: I want you to first make factual statements, then make opinions then make bullet point summaries, have dates, have entity references have your sources for references, all these things. Anyway, so these are things that like I think people who really experiment with structural output have to really care about. But, let's, let's recurse up the stack a little bit. Before we started recording, you mentioned something really cool, which is that there's a lot of engineering that-- inference engineering that goes on when a new model provider releases a new model, right? So let's call it GLM-5.2, Kimi K3. I had previously assumed, especially if it's like, well, GLM 5 to 5.1 to GLM-5.2, like that you've supported them before. Is it that much work?What It Takes to Support a New Open ModelAli [00:11:26]: It's a lot of work.Swyx [00:11:28]: Yeah. Okay. So like, a lot of people, all you guys, right whenever a new model launch like, people rush to say like, “Oh, Hugging Face supports this, Fireworks supports this, Spacetime supports this,” and I'm like, “Yeah, of course we support it.” But what goes into that? What goes intoPhilip [00:11:40]: I think it's more than just support it too, right? It benefits the consumer a lot. Like I think it was with Kimi K2.5 or GLM-5.2 the latest, there was an inference war, right? X provider is at 90 tokens a second. The next day we're at 150. The nextSwyx [00:11:55]: I kinda kicked that off with the GLM-5.2.Swyx [00:11:58]: I wrote a Twitter article about. It got like half a million views,Ali [00:12:02]: Based on being numberSwyx [00:12:03]: YeahAli [00:12:04]: Or it's for something else.Swyx [00:12:05]: Yeah. Which,Ali [00:12:06]: Oh my GodSwyx [00:12:07]: Which then got everyone really excited about, hey, how can we, bend tracks a little bit further and,Philip [00:12:14]: There's a difference between support the model, as in I can make a token out of this model, and support a model, as in I have a production-ready API from this model.Philip [00:12:26]: Getting to the point of I can make a token out of this model is not that hard because generally the, open source inference engines, vLLM, SGLang of the world oftentimes even receive weights ahead of time, maintainers do, or the people making the model merge PRs to ensure support. So you generally can, just get it working on the standard open source stack without too much pain in most cases. The challenge is, every inference company is gonna have own proprietary stack. Some open source components, some in-house stuff. And for any arbitrary model, there's going to be some new stuff. Sometimes you get lucky, like K, two five to two six was, like, pretty similar.Quantization, Speculators, and Production ReadinessAli [00:13:16]: Yeah. It was pure continued post-trainingPhilip [00:13:18]: YeahAli [00:13:18]: If I remember correctly.Philip [00:13:19]: Even in those cases, there's still stuff you have to do. You have to redo the quantization work. You're taking the model from. Generally, these models are not released in NVFP4, and we want them to be in NVFP4 for maximum Blackwell compatibility. So we have to perform that quantization, and, calibrate the quantization to make sure that we're not causing any regression in the model's intelligence. And then we also have to train the speculator, as we've talked about. Generally, we have. We have ZDR, zero data retention on our model APIs, so we don't know exactly the traffic that people are sending us, but we know what's popular. We know that coding use cases are popular. We know that agents, agentic use cases are popular. So we can get public data sets that are representative of that traffic and train general speculators. Now, with speculators today, you need to train the speculator using the base model itself because you're getting hidden states out of the model from running inference on these specific prompts, and that is the training data you use to create the speculator. So there's that process which you need the real model weights for. And then there's of course just the process of, standing up all the infrastructure behind it, loading all this stuff, testing it. And then when there's a new model with a newer architecture, I think that, like, the DeepSeek models tend to be the most challenging as they have, like, the most novel architectural stuff going on, model after model. But every new model has something. Kimi K2 had. Oh, sorry, GLM-5.2 hadAli [00:14:53]: Sparse attention.Philip [00:14:54]: Yeah,Ali [00:14:54]: YeahPhilip [00:14:54]: the DSA.Ali [00:14:55]: Right. Which is brought from DeepSeek.Philip [00:14:57]: Yeah. AndAli [00:14:59]: So you can copy-paste then?Philip [00:15:01]: It kindAli [00:15:01]: I don't know how this works.Philip [00:15:02]: So, like we had to, like, build support for that into our runtime. And you're right, like it is really interesting the way that all of these open source labs borrow from each other. For example, like GLM-5.2 doesn't have vision. So something that, Haley, a guy on our team, if we could take a look at this, he, like, grafted the Kimi vision encoder onto GLM-5.2.Retrofitting Vision into GLM-5.2Ali [00:15:27]: We'll be training the projector.Philip [00:15:28]: Exactly. So if you think about, like, the encoder, there's the encoder, which is the part that looks at the image and turns it into latent information, and then there's the projector which likeAli [00:15:38]: You can say latent space. It's okay.Philip [00:15:41]: And then there's the projector that maps it onto, the model itself, and then there's the model weights. You don't wanna mess with the model weights because you run a chance of making the model dumber at something else for the purpose of giving it vision. So instead, Haley started with just a projector, which is only a handful of millions of parameters.Ali [00:16:02]: That would be, yeah.Philip [00:16:02]: Yeah.Ali [00:16:03]: Can you show the training one?Ali [00:16:04]: Like the way it groksPhilip [00:16:05]: YeahAli [00:16:06]: Very interesting.Philip [00:16:06]: And maybeAli [00:16:07]: That right therePhilip [00:16:07]: Maybe Ali, you should take it from here. You've got a betterAli [00:16:10]: Ooh, double the sandPhilip [00:16:11]: Understanding of this than I do.Ali [00:16:11]: Yeah. You can see, like, he. The way he trained this is really cool. At the beginning, he was training it using just like, “Here's a picture of a mountain. Can you describe what's in this mountain?” And that caused it just like the first, learning walls. Like here you can see this all we're trying to teach it is to translate the encoded. Like it's already taken the encoder from Kimi K. It's taken the image. It'Philip [00:16:31]: Yeah. FrozenAli [00:16:31]: FrozenPhilip [00:16:32]: With adapter.Ali [00:16:32]: Exactly.Philip [00:16:33]: Yeah.Ali [00:16:33]: So the brain is frozen and the eyes are frozen. It's just we're tryingPhilip [00:16:37]: AlignAli [00:16:38]: Interconnect between the eye and the brain, right? So the projector. And so you take the tokens and then he's like, “Oh, can you describe what's in this image?” And he's like, “Oh, it's a mountain,” or it's a person or it's a human, whatever the case is. But that didn't cause complete understanding. So he changed it such that every image was associated with a data set of questions. Like, does this image have a white male? Does this image have birds in the top corner? Does this image have a scientist in it? All of that stuff. And it would have to answer questions correctly. And using not just training on describing an image, but being able to answer question, another question, answer over time. Like you can see the grokking, which is like genuinely insane, that retrofitting vision into a large LLM can learn to that extent. And even for images that it doesn't perform well on, for instance, if you ask it a picture of like Stephen Hawking, “Who is this?” Maybe it doesn't get it, but it will say something like, “This is Albert Einstein.” Like it still understandsPhilip [00:17:25]: Close enoughAli [00:17:26]: That this is a scientist who is a man who has, some significant achievements, all that stuff. So that's like really cool.Philip [00:17:32]: Yeah. So, we've covered Hao Tian before, who the author of the LLaVA paper that did this, a while ago. And I think that's very foundational work for anyone who hasn't done vision work before.Ali [00:17:41]: Same with the CLIP and MetaCLIP, where you go from just captioning to building out questionsPhilip [00:17:47]: RightAli [00:17:47]: Off the image and how much better you can get performance.Philip [00:17:50]: Right. Right. Right. Yeah. But what's, what's so exciting about this is if you look at a model like this. Now, this is a little bit more of a research project. It's not. It got to 56% on MMLU Pro, I think. So not quite frontier. But if you're running this model, you haven't suffered any loss on your GLM-5.2 quality. If you don't have an image, it'll just behave exactly the way it used to. And ultimatelyAli [00:18:14]: Which in the inference code you literally do not include the other part, right?Philip [00:18:18]: Yeah. You would just skip the encoder if you don't have an image input.Ali [00:18:22]: Okay.Philip [00:18:22]: Just confirming.Philip [00:18:23]: YeahAli [00:18:23]: Does it affect a lot on the overall inference side? Like you're not adding much, you're adding a very small vision encoder. These are typically likePhilip [00:18:30]: They're super fineAli [00:18:31]: Less than a billion parameters, right?Philip [00:18:32]: Yeah. It's, - There's a little bit less standardization among vision encodersSwyx [00:18:37]: YeahPhilip [00:18:37]: So the support matrix can be a little bit, sparser. But overall, yeah, it's a pretty, it's a pretty minor component of the overall system. And ultimately what you get out of the system is all of a sudden you have Kimi Vision, GLM weights, and DeepSeek attention all in one model.Open Source Model Grafting and Franken-MergesPhilip [00:18:56]: And that's, I think, a lot of the power and beauty of open source, is that you can take all of these different components and combine them together into a system that's better than anyoneSwyx [00:19:05]: YeahPhilip [00:19:05]: Can be individually.Swyx [00:19:06]: People used to say that you would also do Franken-merges where you would take likePhilip [00:19:10]: YeahSwyx [00:19:10]: Layers from each model.Swyx [00:19:11]: Does anyone do that anymore?Ali [00:19:13]: Well, to your point previously when you were mentioning like, the work that goes into supporting a model when it first comes out, like GLM-5.2 or MiniMax M3 or whatever the case is. Sometimes you do have to like, you do have to switch out some things. Like, for instance, the MiniMax M3 head uses full attention, and with full attention you end up with this like insane bottleneck in spec dec ‘cause you're doing auto-regressive token generation for three tokens, and you're doing this like N squared over all of the tokens that are in your sequence. Your KV cache is like very large because it's not sparse, it's not top K. So we find it better to like, okay, we're gonna replace this, we're gonna replace this layer with a layer from another model that's using like GQA, for instance. And then just with the right training, you can get it to have the same acceptance rate. So it is very possible to retrofit layers from other models and very much needed. If a layer is like inefficient, the training just becomes the challenge, like how do you ensure that you train it properly? Which again to your earlier point is like the mesh between training and inference. As in like you need very good training in order to do fast inference. That's like, I feel like more and more becoming true.Swyx [00:20:21]: Yeah. Anything else on the support side when you say like get it to fully production ready?Loop Detection, Race Conditions, and Non-DeterminismPhilip [00:20:26]: Yeah. I think that there's also a question of just, we can test a model to a pretty extensive degree, but we're trying to get it out quickly and then you see a bunch of other people test it and you get interesting results. There was an issue with, GLM briefly where we had some like mode collapses where it would just output the same token over and over again for certain prompts on certain temperatures. Like once you expose an endpoint to the real world, there's going to be, so many more varieties of things given to it that you're able to, discover and patch things. So it's not just a, day zero process, it's then like for the first week, for the first month, if a model remains popular, like how do you both fix bugs and then continue to push the envelope on performance?Ali [00:21:21]: What do you mean you don't want your model outputting S?Swyx [00:21:24]: Is there loop detection on that stuff, by the way? It still happens like quite a lot, which is surprising.Ali [00:21:30]: We have like we, in our endpoint, like if a model was to output the same token like four plus times, we just cut the generation. We say like, “Oh, sorry, this-- Like try again,” or like we will reprocess the request. ‘Cause we know then, like if it, like if, yeah, it's four times the same token, it's probably collapsed.Swyx [00:21:45]: Yeah. Is there a way to opt out in case I really want that?Ali [00:21:48]: You want that?Ali [00:21:50]: I think there's a way that we have to handle it. I'm not exactly certain, but I feel like in certain models, like when they output something like you can imagine, like a table for instance, and so they want, they wanna draw like 12 dashes and 12 dashes. Yeah, I think there's a way for that to happen. I think we only do it on certain tokens. Like we exclude certain special characters.Swyx [00:22:07]: Yeah.Ali [00:22:07]: So we only do it on like certain like S is the most common almost. GLM-5.2Swyx [00:22:11]: OhAli [00:22:11]: And I think it was DSV 4 as well. Like you'd just have like looping issues where like you literallySwyx [00:22:17]: ItAli [00:22:17]: Just have like S.Swyx [00:22:18]: Yeah. Is there a special, something special about S? No, just randomlyAli [00:22:21]: It just seems to be the one token involved.Swyx [00:22:23]: Yeah. And it'Philip [00:22:24]: Is thereSwyx [00:22:24]: And it's only temperature 0Ali [00:22:27]: NoSwyx [00:22:27]: Even at other temperaturesAli [00:22:27]: Even at like 0.9 or whatever, it will still, it will still collapse.Swyx [00:22:30]: That's weird, right?Ali [00:22:30]: It's, it is an inference problem to be honest, like a software problem. Like oftentimes, the image you run will-- like NVIDIA will release an image for instance, and if we will upstream the changes from their latest TensorRT-LLM image into our stack, we'll find that it fixes it. Or oftentimes this will only happen in an inference engine that you're using like SGLang. But if you were to switch to vLLM, that isn't the case. So it seems to be like an extremely like deterministic software issue and not really a model issue. It's not like a weights problem. Like I'- we'll say like, “Oh, it's a problem with the quant. We did PTQ wrong,” right? But that isn't, that doesn't make sense because the same weights used with a different inference engine does not repeat the problem. And sometimes it's, the kernels that are being used in the backend have like these very subtle sometimes race conditions, where if you were to use this model hosted on one cluster, you will never get this problem.Swyx [00:23:19]: Oh my God.Ali [00:23:19]: But if you host it on a different cluster, you will. And the reason is the KV cache transfer from a node to node in that one cluster is using a slower interconnect than the node to node in another cluster. So that exposes the race, whereas in another cluster it doesn't. So then you end up just like, okay, this model is not gonna be hosted on this cluster. We're gonna host it on, another cluster because that cluster exposed that problem. But then it ends up with like, okay, is it the software? Is it the model weights or is it the hardware?Swyx [00:23:42]: There is a thing about this with temperature 0 still not being deterministic, right?Ali [00:23:46]: Right.Swyx [00:23:46]: Mostly because of hardware. Even at temperature 0 same model, you won't always get the same output.Swyx [00:23:52]: Even-- But I'm surprised by the race condition one because, I thought PyTorch was a graph that like guarantees that you at least, execute things in the right order.Ali [00:24:02]: Well, yeah, true. Like I'm not, I'm not saying that there is. Like well, you have things like PTL optimizations where like you can start a kernel before the end of the previous kernel, and that's like ‘cause you want to do that because there'sSwyx [00:24:12]: It's like pipeliningAli [00:24:12]: Expense. Exactly.Swyx [00:24:13]: Yeah.Ali [00:24:13]: But it'- But you don't do it cleanly. Like you overlap a little bit of the execution. No, it is very possible that the kernel itself, like that one block that is supposed to be running in this instance of time, that kernel itself has a race condition. For instance, like a missing barrier. Like often if you're designing a kernel and you want it to make it to be very fast, if you don't test it extensively, you'll, you'll have certain threads access data points from registers before they've been written to by other threadsSwyx [00:24:36]: YeahAli [00:24:36]: For example, because like your barrier is wrong or your synchronization was wrong. But yeah, like the testing itself is very difficult in those like, andSwyx [00:24:42]: And there's no like borrow checkerAli [00:24:45]: What does that mean?Swyx [00:24:46]: Like Rust. Like the. If you're trying to have like memory safety It sounds like a comparable problem.Ali [00:24:52]: Well, yes, but you're working in CUDA, right, NVIDIA GPUs. Like- You just need a higher level language like modular Maybe that's what modular is supposed to do. I don't know.Quantization Quality and Vendor FidelityVibhu [00:25:00]: How do you see keeping quality of the model? So you talked about all these steps of, okay, you gotta do quantization, train your own speculative decoderAli [00:25:07]: RightVibhu [00:25:07]: Run on different hardware. Looking at other model providers, okay, you kicked off a inference speed race on the consumer end. What goes into keeping quality the same across them, right? Sure, you can run benchmarksAli [00:25:22]: YeahVibhu [00:25:22]: But, like, how do you determine how much quantization are there standards? What goes intoPhilip [00:25:27]: There's a few things on quality. Most inference optimizations are lossless. KV caching, for example. You are just recomputing or preventing recomputing the same values. Speculation, of course, if a draft token is wrong, it gets rejected. The main lossy optimization is quantization. And that really comes down to, number one, data format, number two, which parts of the model you choose to quantize, which layers, and number three, like doing a lot of calibration on the quantized weights, to ensure that you're preserving all the outliers. There's other tricks that you can do, though. A big one is long context, ‘cause one thing you asked at, right at the beginning is, “Oh, what's gonna happen if I send a 200,000 token request in?” So with a long input sequence, you need to, store a lot more information. You need to process a lot more tokens. And so even if a model has a context of a certain length, you might, as an inference provider, choose to build an API with a shorter context length, and of course a full length one as well. Because if someone doesn't need the full million token context, for example, you can get them better performance. I don't know if that's exactly like quality of the model. The way that I think about quality is to what degree are we faithfully serving the original model? If you think of a golden implementation of a model that performs exactly the way the model is designed to perform, I think of quality as how close are we getting to that, 100% fidelity of the model.Philip [00:27:13]: You can also, of course, think about quality from the training side and how do you push yourself past 100%. But when I think about purely inference optimizations, it's getting faster while staying as close to that 100% fidelity mark as possible. And certainly our standard internally is that, like you should not be able to tell the difference between our API and a, official API. I think Kimi in particular does a good job of vendor benchmarking hereAli [00:27:41]: YesPhilip [00:27:41]: Where they haveAli [00:27:42]: They released an actual vendor benchmark.Philip [00:27:43]: Exactly, yeah.Ali [00:27:44]: ‘Cause they accused, some people, Amazon? There was some provider that was not doing very well on Kimi's benchmark.Philip [00:27:50]: Yeah.Philip [00:27:51]: So, with Reflect we probablyVibhu [00:27:52]: This was a long time ago, right?Philip [00:27:54]: No.Ali [00:27:54]: Yeah, like threeVibhu [00:27:55]: They alsoAli [00:27:55]: Four, five months agoVibhu [00:27:57]: This also happened with, I don't remember which model, but they pulled out quite a few, and then they started a whole chart about this. It might have beenPhilip [00:28:03]: Kimi Vendor Verifier.Ali [00:28:04]: Yeah.Philip [00:28:05]: Yeah.Ali [00:28:05]: Yeah, ‘cause you, ‘cause you'd be pissed, right? Like if you'Philip [00:28:07]: Yeah.Ali [00:28:07]: If like if I'm a consumer and I'm using like Amazon's endpoint for instance, and I've used Kimi and I'm like, “Oh my God, like this is bad,” I'm not gonna say, “Oh, Amazon quantized the model in a bad way.” I'm gonna say, “Oh, Kimi sucks.” Right?Philip [00:28:17]: Yeah.Ali [00:28:17]: So it seems like that makes sense.Philip [00:28:19]: Yeah, they care. They care.Vibhu [00:28:21]: Justifiably.Ali [00:28:21]: Yeah, justifiably.Vibhu [00:28:22]: This is probably a stupid question, but just checking, has anything improved from main quantization?Philip [00:28:28]: Yeah.Vibhu [00:28:28]: Like, is quantization always strictly worse?Ali [00:28:30]: Well technicallyVibhu [00:28:32]: NoAli [00:28:32]: It's a lossy. QuantizationPhilip [00:28:33]: YeahAli [00:28:33]: Is a lossy, it's a lossy implementation.Philip [00:28:36]: Speed improvesVibhu [00:28:36]: Speed improves.Ali [00:28:37]: It the number, likeVibhu [00:28:38]: No, I' always look for inverse scaling laws.Philip [00:28:40]: Yeah.Ali [00:28:40]: Yeah.Vibhu [00:28:40]: This is something I learned from Noam Brown, where like things that normally act in one direction sometimes do.Philip [00:28:45]: Well, technically when you run a benchmark, because these models are deterministic, sometimes your,Ali [00:28:52]: YeahPhilip [00:28:52]: NVFP4 quant is like, two basis points higher than yourAli [00:28:56]: No, it's noise. It's noise.Philip [00:28:57]: Yeah, exactly. I'm like, yeah, it's, it's within. That's why I always say within margin of error.Philip [00:29:01]: And I stopped saying that because everyone assumes that what is, well, within some margin of error, we're barely inside of that to the worst, so we're saying. But yeah, sometimes it's just like, gives you a higher output score. But like Ali said, that's noise. To my knowledge, you're not necessarily making the results better. You're just trying to, again, like keep your fidelity as close to 100% to the original model.Layer Selection, KL Divergence, and Better QuantizationAli [00:29:27]: There is, to your point, research that we did on MP. I don't know if you are able to pullPhilip [00:29:31]: YeahAli [00:29:32]: A tweet we did. One of our research interns, Joshua, I think it's a tweet on how we have 20% better quantized GLM-5.2 than NVIDIA. Essentially what we found throughout like this month research is, okay, quantization is a lossy. It's. You're compressing the data from, occupying 16 bits to occupying, four bits, for instance. And so you're losing some information, and you're trying to minimize that. And so when I say that I'm gonna quantize the model, my job becomes how do I find the layers that I can quantize, and how to find the layers to not. For instance, with image models, I don't quantize modulation layers, and I don't quantize out projections because those two are. Like out projection is what you see as the user. Modulation is what the model sees or understands. Right, exactly. And so to his paper, do you have the. It doesn't have the. Yeah. It's a long paper. I don't know if I can findVibhu [00:30:25]: If there's a part to search or it's probably in the thread.Ali [00:30:28]: It's probably in the thread.Vibhu [00:30:29]: Yeah.Ali [00:30:29]: But the long and the short is it is very possible that quantizing more of the model makes the results. Like if I have a model that I quantize layers one, five, and 10, and another model where I only quantize layers one and It is possible that the model in which I quantized more information is going to perform better because the quantization errors have canceled out. And so what Joshua showed in his mathematical proof where he had like a verifier in, is that you can predict which layers are going to have quantization errors that will cancel out with each other, and you choose to quantize those layers. And so the result of doing this mathematical quantization is you end up with a model that's 20% more quantized than another provider, so you get 20% more throughput of it because there's more layers than running an NVFP4, and your quality is better than that other quant because the layers that you chose to quantize have their errors cancel out, like one layer skewed to the right one layer skewed to the left, one layer skewed to the right. Your final logits distribution is more similar to the original distribution of the model, so you have better fidelity. And so the way we proved this was with KL divergence. So instead of just scoring on the benchmarks, we scored the KL divergence between the logit distribution of the quantized model and the logit distribution of the original full precision model, and we showed that with this technique we get. If your probability distribution on the logits which token it wants to select is more of the same as the original model, you're probably gonna end up staying true to the original model. So yeah, so it seems like previously before this, it seemed like the industry was, well, the more you quantize, the worse it's gonna be, ‘cause the more loss you introduce. That's not exactly, not necessarily true. So yeah, doesn't improve it, but can cancel out.Philip [00:31:57]: I think it might be this, but reminds me a good bit about pruning where you can prune off certain layers.Philip [00:32:03]: But very interesting. Didn't know this was a whole paper you guys put out.Ali [00:32:06]: It's. Fun fact, it was originally 72 pages, this paper, and then we decidedPhilip [00:32:11]: WowAli [00:32:11]: We can't tell. We couldn't release it. So it's now 45.Swyx [00:32:15]: Still 39 pages, so very substantive. We talked about evals and all these things and, like what's possible in terms of speedup? Like it's like probably like the numberInference Speedups and BenchmarkingSwyx [00:32:25]: Thing that people do wanna care about, and it's something that you wrote about in your post. Like official API is 70 tokens per second, and you push it up to 90. Is that like a normal thing?Philip [00:32:36]: So what's cool about working in inference, the reason that I think inference is going to be a useful place to do engineering for a long time, is that if you look at highly optimized domains like, say, finance, if you're in finance, you measure how much better you got in basis points. It's like, “Oh, I got five basis points better, like twentieth of 1% better,” that's huge news because everything is so optimized. When we publish optimizations, it's 20%, it's 100% it's 200%. So there's still probably like a lot further to go, honestly. Like you'll, you'll know that inference is pretty much solved when researchers start publishing about how they got 1% faster at something.Swyx [00:33:19]: Which by the way, because I am from the finance background, in the ‘70s, that was the margin at the time. When you did quantitative finance research, you would findAli [00:33:27]: And like 20%, tens of percent.Swyx [00:33:29]: That's. Yes.Philip [00:33:29]: Yeah.Swyx [00:33:30]: And now it'Philip [00:33:31]: Tiny fractionsSwyx [00:33:32]: For those people interested, look up Andrew Lo's paper. He had a really interesting illustration of quant, stat arb, distribution, narrowing down from like those kinds of 20% differences in the ‘70s, down to nothing today, which is very cool.Philip [00:33:48]: Exactly, and we're at the beginning of the same type of thing. Now benchmarking is hard. I think anyone will tell you that, and benchmarking provider speeds is hard because there's so many variables that go into it. What hardware are you using? How much load do you have on the system? What's the exact nature of the prompts and input and output sequence lengths? All that stuff. But overall, when you start stacking these improvements, you're looking at multiples. You can look at it. The most common form, of course, is TPS, tokens per second, which is bad naming by us in the industry, ‘cause there's two tokens per second. There's tokens per second, the throughput number, and the latency number.Ali [00:34:31]: TTMT, yeah.Philip [00:34:32]: Like total tokens per second out of the, out of the GPU as a throughput number. Most people only care about tokens per second as the latency number, which we should call ITL, intertoken latency, but we don't.Philip [00:34:44]: Anyway, so you can imagine a standard API without many optimizations for a 1 trillion parameter model operating somewhere in the 30 to 50 tokens per second range for reasonable traffic profile. And we generally see the goal of, pushing to 10X that. But, not necessarily day zero, but by stacking enough optimizations, if you have, say like four optimizations, each of which doubles performance. Or sorry, three optimizations, each of which doubles performance, then you stack that up, that's an 8X gain. That's the order of magnitude that we're working with in this space. We're trying to make things substantially faster, not just go from like 70 to 90.Swyx [00:35:38]: Are you saying you've. You have done that?Philip [00:35:40]: So let's say you have as a reasonable baseline, 30 or 40 tokens per second. You can achieve 10X that. So like on GLM-5.2, if you run it unquantized, perhaps on H100s even, and you're just using an off-the-shelf inference engine with no particular optimizations, no speculator, nothing extra around like KV routing, no disaggregation, you're, you're probably, yeah, looking at that like 30 to 40. You think that's like a reasonable baseline?Swyx [00:36:12]: Right. Right.Philip [00:36:12]: To get to something like 10X, there's a lot of trade-offs that you're making. If we're running at more like a 300, 400 tokens per second range, you are using the best hardware possible. You have a optimized speculator. You have done all of your quantization work. You are Seeing a pretty high cache hit rate. You are running with a reasonably small batch size and a parallelism configuration that is tuned for latency versus throughput, but it is possible. So the spreads that you see if you, like, go on artificial analysis or you go on OpenRouter and you look at, the worst provider to the best provider, oftentimes can hit that range. 10X is of course very aggressive. It's oftentimes maybe more of a four to six times improvement. But that's the performance that makes us really excited, is when we can get these huge gains, not just go from 70 to 90 tokens.Stacking Optimizations: NVFP4, Speculation, and DisaggregationAli [00:37:19]: It's also, like, hardware dependent. Like, ifPhilip [00:37:20]: YeahAli [00:37:20]: If you have a thing where you're serving it on just, like, a node of H100s and then you throw, like, you shard the model across, like, four nodes of B200s. Like, you can definitely increase the speed with just throwing more hardware at it. Like, normalizing for the same exact hardware and the same number of GPUs.Philip [00:37:35]: Yeah. Then you're looking at, like, a two to 4X improvementAli [00:37:38]: Right. RightPhilip [00:37:38]: Depending on the inference optimizations. So yeah, it's. Some of it's, what's the call, and some of it's who's the driver.Vibhu [00:37:46]: If you break down the two to 4X, say the example is run GLM-5.2Ali [00:37:51]: YeahVibhu [00:37:51]: On B200sAli [00:37:53]: YeahVibhu [00:37:53]: Single node, right? What's, like, the cost trade-off for effort to get, like, the last bit of juice out versus what should people just think of, right?Ali [00:38:01]: Spectre quantization. Yeah.Vibhu [00:38:03]: Spectre quantization.Ali [00:38:04]: That's, that's, that's like 95%. LikeVibhu [00:38:06]: And how far does that get you? And how easy is that for the average person to do? So say right I wanna throw the weights of GLM-5.2 on a node of B200s, how easy is it to find speculative decoder- decoder model or already quantized model? How much work goes into it?Philip [00:38:23]: If you're doing it up front, it's quite a lot of work. If you're doing it today, there's going to be people who have published things that you can just, you can just grab some NVFP4 weights. You can grab a speculator. Yeah, if we're thinking about, like, what are the 2Xs we're stacking, going from, BF16 to NVFP4 is, it's not quite a 2X, right? It's like. I think it's about, like, 30 to 40%, from 16 to 8, and then another 30 to 40% multiplied from, 8 to 4. So that doesn't quite get you a 2X, but, like, roughly a 2X. Speculator, roughly a 2X. Disagg on top of that if you're able to get enough hardware and put enough traffic through it, another roughly a 2X. And then you add in some, double-digit percent increase from having just a better runtime with, the latest kernels and stuff behind it. And that's how it stacks up.Ali [00:39:21]: YeahPhilip [00:39:21]: So building each of those, like, building the, quantized weights is, for someone who really knows what they're doing, hours to days of work. Building the speculator, again, like, hours to days of work. And the, disagg setup, hours to days. Well okay, but like once you haveAli [00:39:39]: Once set up. Once set up. YeahPhilip [00:39:40]: Yeah, getting disagg working for the first time, I'm saying, of course, is very difficult.Philip [00:39:44]: The marginal implementationAli [00:39:48]: Like, if you're just grabbing, like if you are a person, like just a normal consumer who has access to, like, a node of B200s and you're wondering, “How can I just host it myself?” You don't need to quantize the model yourself. There's always gonna be, like, an open source quantized checkpoint. NVIDIA's gonna push one out if no one else does. You. Usually, the providers will have their own spec dec that they've trained as well. You don't need to train your own spec dec. You can just use that as well.Philip [00:40:09]: Yeah. Like, GLM-5.2 has its own MTP.Ali [00:40:13]: Right. Right.Vibhu [00:40:14]: What's multi token prediction?Philip [00:40:15]: Yes.Ali [00:40:16]: I'm justVibhu [00:40:16]: Can you explain that?Ali [00:40:16]: I'm just an expert.Ali [00:40:18]: I can do it for you in case I get it wrong?Vibhu [00:40:20]: No.Vibhu [00:40:21]: Yeah, you should correct if we're wrong, but their multi-token prediction can be used for self-speculative decoding.Ali [00:40:27]: I'm not sure. I'm not gonna correct that.Vibhu [00:40:28]: Okay. I'm semi-confident in thatAli [00:40:30]: Okay. YeahVibhu [00:40:30]: But someone can check. But it's useful to paint the story of, okay, not just the average person, but say a company wants to switch from serverless inference I wanna throw this up on. I wanna rent some GPUs, throw it up. These are the steps you take to do significantly faster than just put it behind vLLM.Ali [00:40:48]: Right.Vibhu [00:40:49]: I was waiting for a mention of Dynamo.Vibhu [00:40:51]: I feel like, that's supposed to be the baseline that you measure against.Dynamo, KV Routing, and Disaggregation ToolkitsPhilip [00:40:55]: I would think of Dynamo as less of a box system and more of a toolkit for building with. So when we talk about doing aware routing, when we talk about doing KV offloading, when we talk about doing, PD disaggregation, Dynamo fundamentally is. By the way, Dynamo is an open source library from NVIDIA.Ali [00:41:17]: We've done a pod with KylePhilip [00:41:18]: OkayAli [00:41:19]: Kyle Cranin.Philip [00:41:19]: Cool. So then your listeners know then that it supports all the different inference frameworks. And it is multi hardware, which is interesting.Ali [00:41:28]: But it's just a router, it's not like an optimizer layer.Philip [00:41:30]: Yeah. All it does, like, what Dynamo is good at, it is a library for moving information around your cluster, around your hardware. So if you have, KV cache on one place and you need it to be somewhere else, Dynamo coordinates NIXL for you to move that around.Philip [00:41:49]: That doesn't mean that, like, out of the box, you just say, “Pip install Dynamo,” and then you get, like, a massive performance speed up. It's more of a developer toolkit.Ali [00:42:01]: Yeah. I would have said it would. It comes with a set of defaults that you can then swap out.Philip [00:42:06]: It does. If the industry at large, I think, was, like, rolling out all of these deployments, standard, then I think it would be, like, a credible baseline. But, we've got to, we've got to benchmark against, like, what we're seeing in the wild.Speculative Decoding Methods: Medusa, EAGLE, n-Gram, and Spec-SpecVibhu [00:42:23]: I did wanna talk a little bit more about PD disagg, because that is probably, like, number three after quantized and speculative decoding. In your book though, I was just gonna pull out the book.Philip [00:42:31]: Yeah.Vibhu [00:42:32]: Like section 522 on Medusa, 523 on EAGLEPhilip [00:42:35]: YeahVibhu [00:42:36]: 524 on gram.Philip [00:42:37]: It's 55, would be disaggregationAli [00:42:42]: Yeah. Well, no, I just wanted to dwell a little bitPhilip [00:42:44]: YeahAli [00:42:44]: The other. Like, so what do you choose to include? What do you choose to not to include? Because there was all these other techniques.Philip [00:42:51]: Yeah.Ali [00:42:51]: Are these still relevant? Because I think they came out, like, a year and a half ago maybe.Vibhu [00:42:55]: Medusa is quite old.Philip [00:42:56]: Yeah, Medusa's old.Ali [00:42:58]: It was old.Vibhu [00:42:58]: But is it in the book as a good, here'sPhilip [00:43:01]: BaselineVibhu [00:43:01]: Baseline vanilla understand it?Philip [00:43:02]: Like you should know this.Vibhu [00:43:03]: Like I read the paper, I'm like, “ it makes so much sense.”Philip [00:43:05]: Yeah.Philip [00:43:05]: So with the book, I had a couple goals. One was to give people just a working vocabulary for the space as a whole, and the other was to give them some intuition about how each of these techniques works. As I mentioned in my AI Engineer talk, which is the first public addendum to this, the speculation space has moved much faster than everything else. So yeah, even at the time that I wrote the book Medusa, I very much included as a way for people to understand how the space evolved rather than what the most modern technique is. And now of course, there's DFlash, dSpark. There's, there's newer techniques even than EAGLE, although EAGLE is still very commonly used.Ali [00:43:51]: SpecSpecta.Philip [00:43:52]: Yes. Speculative decoding.Vibhu [00:43:54]: What canAli [00:43:56]: Oh, it's a paper by Tri Dao and it's like, it's doing speculative decodingVibhu [00:44:00]: HuhAli [00:44:01]: For the speculative decoder.Philip [00:44:02]: Oh, in spec- oh my God.Ali [00:44:02]: It's literally just an another. It's like, yeah, that's the most simple way to explain it, and it seems like he got trivial speed ups there. But it seems that the complexity with training, it's almost like in our mind at least, it's almost as complex as training GANs. Like it's like a very delicate balance and oftentimes you, it's just but yeah, it's literally speculative decoding on speculative decoding.Vibhu [00:44:21]: Speculative.Ali [00:44:22]: Yeah. We saw this paper.Vibhu [00:44:24]: It's interesting, right?Ali [00:44:24]: Yeah.Vibhu [00:44:24]: I wouldn't even expect it to be very particular to train, I wouldAli [00:44:29]: Right.Vibhu [00:44:29]: The naive part of me is like, okay, train speculative decoder.Ali [00:44:32]: But like, and it makes sense, like the whole idea of speculative decoding is you. It's like, it's like almost like the iPhone auto predict version but for a normal model, right? Like you're just, you're just, generating three tokens and you're like, okay, I'll do prefill on them. And so you save those three turns for your original model. Now your speculative decoder is doing three turns of auto regression, so why not just have an even smaller model?Ali [00:44:53]: The other question there is what are the size of speculators? So say forPhilip [00:44:58]: Right. It's like a billion parameters.Ali [00:45:01]: Like for MiniMax, it's. Yeah. It's like one layer. It's like one 60th of the original model usually.Philip [00:45:06]: Yeah. I think we should do a paper when we get back to the office.Philip [00:45:10]: SpeculativeAli [00:45:11]: SpeculativePhilip [00:45:11]: Decoding.Ali [00:45:13]: No, it's, it does seem like how, when do you stop? But then it also seems like if you're able to train spec-spec decode for instance, right? Like if you're able to have a small model that is accurately predicts what the intermediate speculator is gonna predict, that is able to predict what the original target model's gonna predict, then why not just use that smallest model directly, right?Vibhu [00:45:34]: Yeah. This isAli [00:45:35]: Like it seems likeVibhu [00:45:35]: Adjacent to the routing problem.Ali [00:45:36]: Right.Vibhu [00:45:36]: Yeah.Ali [00:45:36]: Right.Philip [00:45:37]: The thing with speculators is one of the practical constraints on using them is that you do have to run a small model on the same hardware that you're running the big model on. There is a orchestration and resource competition problem inherent in that, and that is one of the constraints on speculation in general, is that draft tokens cost resources to create and cost software complexity to manage. And so if you have like infinitely recursive speculators, you add in quite a bit of that complexity on the actual implementation within the inference engine as well, not just in the training process.Vibhu [00:46:17]: I was gonna say, I would wonder if you could do similar, like distillation and pruning of, it's the same thing, it's just a model. Can we not just distill a lot of the weights, quantize the speculator, out of my domain? The question that also comes up is, this is all for big server workloads, right? How much of this applies to, say I have this MacBook, I wanna run Gemma really efficiently. Similar problems, not the same?Local AI vs. Data Center InferencePhilip [00:46:45]: Pretty different. I talked to Selo, about this on his podcast a couple weeks ago. The difference between inference engineering for the data center and for production workloads versus inference engineering for local AI, is that we start with fundamentally like different constraints and different goals. With local AI, it's how do I fit this model onto my hardware and then make it less dumb? And with data center influence, it's how do I load this model and then make it less slow? And we care about less dumb, and they care about less slow. But the local AI inference engineering ecosystem, I think has a lot for us to learn from in the data center space. They are experts in various forms of quantization, including dynamic quantization that we just don't touch, in the pruning, in the distillation, in the, layer removal. There'Ali [00:47:42]: Layer removal matters less.Philip [00:47:43]: Yeah. There'Ali [00:47:44]: No one loves pruning really.Philip [00:47:45]: Yeah. Well, but the, but they doVibhu [00:47:46]: Which is surprising, right? But that's, that's a whole different thingPhilip [00:47:48]: Just to fit something on the laptop.Ali [00:47:50]: Right.Philip [00:47:50]: So yeah, it's a, it's an interesting, it's an interesting space. Not necessarily that like their techniques make sense for us to do in the data center, because we have different resources and different goals, but more that the process as well as the openness of that field is something to, admire.Ali [00:48:12]: Yeah. Like to your point, like, certain optimizations that would. Like for instance, Turbo Quantum Sharper, like it made such huge hype on that and we did like a whole deep dive on Twitter and like said, what is it? How does it work? Why is it good or not? And it took off and it was implemented on local devices because your memory bandwidth is so slow on like a MacBook, for instance. But try putting the same thing on like an NVIDIA GPU on a B200 Turbo quant would not be. Like, it would not be used. Like, NVIDIA - Like, NVIDIA made it clear that this is not a good optimization, and we've seen it firsthand where the overhead of doing dequantization, quantization of, in the kernel itself with turbo quant kernel, each end is much slower than the time that you save from doing the bandwidth. ‘Cause on the B200s, you have like 3.5 terabytes per second. You don't need decrease the storage that much. You don't need to do, FP4 KV cache. You don't need to use a requant. There's, there's, there's better optimizations to be made. But on Edge devices, it's extremely important, it's extremely useful. So, seems to be, like, different optimizations there, but then they're all uniquely combined with like all you wanna quantize the model, you wanna do speculative decoding, like certain common prefixes with bothPhilip [00:49:18]: Principles.Ali [00:49:19]: Yeah, exactly. Exactly. Exactly.Philip [00:49:20]: They also do a lot of work on, model parallelism, especially over, heterogeneous topology, where you have, some sparks and they are wired together with, Ethernet, DGX sparks.Ali [00:49:35]: Yeah, this is the Exo Labs guys.Philip [00:49:36]: Yeah. You have, a nu

No episódio de hoje do Check-up Semanal, o Dr. Ronaldo Gismondi, editor-chefe médico do Portal Afya e do Whitebook, comenta os principais destaques recentes em Infectologia, com foco em HIV, infecções por Clostridioides difficile, bacteremias por Staphylococcus aureus, profilaxia antifúngica em transplantes e resistência bacteriana. Falamos sobre o uso do lenacapavir em pessoas vivendo com HIV multiexperimentadas, a nova diretriz da AGA para infecção por C. difficile, as evidências que posicionam a cefazolina como alternativa às penicilinas antiestafilocócicas nas bacteremias por MSSA, as recomendações da IDSA para profilaxia de aspergilose invasiva em transplantados e as opções terapêuticas para infecções por bacilos Gram-negativos produtores de metalo-beta-lactamases (MBL).Leia na íntegra os artigos mencionados hoje:Lenacapavir como opção para o tratamento de PVHIV multiexperimentadosInfecção por Clostridioides difficile: nova diretriz da AGA 2026Eficácia e segurança da cefazolina no tratamento de bacteremias por MSSAProfilaxia de aspergilose invasiva em transplantes: o que diz a diretriz IDSA?Infecções por bacilos Gram-negativos produtores de MBL: aztreonam-avibactam ou ceftazidima-avibactam + aztreonam?

It's the last day of July why not squeak in a little Christmas? We are unboxing two things this week first, we have @sohi_studios_ Gram glass with their extra thick, heavy wall beaker bong which dare I say is close to indestructible. Dave will be bringing a bong for everyone and you know what that means you get a bong !you get a bong! you get a bong ! When it comes to bongs functionality is the most important part. If you have a bong that doesn't hit right it has no future but if you get one that hits just right, that could be your daily driver for years check out their site or Gramglass.com or their Instagram pages @gramglass.labs and you will see right away that these guys are thinking along the same path of functionality, ease of cleaning and the glass is thick as shit! We also have the @kaktuscutter electric Herb grinder to check out with 3 different models theirs one for everyone. The K1 portable, the K2 slightly larger but still portable and the tabletop K5 which is good for extra large runs and mobility compromised individuals check out all the other storage options and accessories on their website kactuscutter.com Finally when you think of Christmas you gotta think about Thailand or at least dream your there. Our Guest Tian has living that dream for years from working on the legal Us cannabis market to educating People in Thailand how to navigate and serve the public our favorite herb. Here has been a reversal in the cannabis market with a ban on rec and the market going full on into a medical framework. Tian @greentlife is about to embark on a national wellness tour we will get the inside scoop on that and more. So get that @puffco charged your @jerome_baker bong Clean with some ice

Last year, acclaimed singer-songwriter Ebony Lamb and her partner, fellow musician Gram Antler, packed up and moved to Japan for a three-year stint. A real loss for the New Zealand music scene... but fortunately they're back home on a nationwide tour. The tour also celebrates Gram Antler's debut album - 'Novel Lover', with the pair sharing the stage for a show that blends songs from both of their catalogues. And that's exactly what they're bringing us today for NZ Live... a selection of songs from Ebony Lamb and Gram Antler.

Podcasting has become one of the most effective ways to build authority, connect with your audience, and create long-form content that can be repurposed across multiple platforms. Whether you're a business owner, coach, educator, or simply passionate about sharing your expertise, starting a podcast can help you expand your reach and establish yourself as an expert in your industry. While launching a podcast may seem overwhelming, breaking it down into manageable steps makes the process much more approachable. This episode talks about everything you need to know before you hit record. Resources: The Meeting Place Membership Rock The Reels 1:1 Coaching Free Client Welcome Guide Additional Trainings and Resources Connect with Brittni: Follow me on the Gram - @brittni.schroeder Join my Facebook Group  Visit my website Subscribe to my Youtube You can find the complete show notes here: https://brittnischroeder.com/podcast/how-to-start-a-podcast-the-complete-beginners-checklist/

Laylee Emadi, coach for speakers and educators and founder of The Creative Educator Conference, joins Brittni to unpack what it really takes to build a career around teaching and speaking. They dig into the difference between professional speakers and industry professionals who speak, how to strategically pitch and get booked, and the key pieces every aspiring speaker needs in place before reaching out to organizers. Laylee also shares her biggest entrepreneurship pet peeve and the simple pleasure she can't do without. Connect with Layee on Instagram Laylee's Episode Freebie Resources: The Meeting Place Membership Rock The Reels 1:1 Coaching Free Client Welcome Guide Additional Trainings and Resources Connect with Brittni: Follow me on the Gram - @brittni.schroeder Join my Facebook Group  Visit my website Subscribe to my Youtube You can find the complete show notes here: https://brittnischroeder.com/podcast/laylee-emadi-coach-for-speakers-educators/  

Today's episode is a WILD ride! Lol we're a bit all over the place as we share what we've been up to, updates on our new Bible study AND our new book, our latest beef about Kenz's new company, our monthly favs, and of course some of what God has currently been teaching us! BUCKLE UP because it's a FUN one to say the least! Thank you to today's sponsor! Hiya Health... our favorite vitamins + more to give our kiddos! Get 50% off your first order using our exclusive link! Grab our NEWEST Bible Study... Freedom Looks Good On You! It's officially out today! Grab it here on Amazon or here for 30% off + free shipping! Pre-order our book book... This Wasn't The Plan! Follow us on the Gram: @forthegirl___ Our Book Recs: Being with Jesus by Jim Branch The Knowledge of the Holy by A.W. Towzer Learn more about your ad choices. Visit megaphone.fm/adchoices

Improv is Dead Survey SpooktacularClaire Favret joins us to chat'n'prov about sexy chefs, late life crushing, and the incredibly talented and very athletic Banana Ballers. Follow Claire on the Gram!Follow Downpour for upcoming shows!Support the pod and join our Patreon for bonus scenes, our entire backlog, and even more premium content!Just easing into Improv is Dead? Check out our Starter Platter and Best of Playlists on Spotify!Hosts: Tim Lyons⁠ and Damian Anaya

47e6GvjL4in5Zy5vVHMb9PQtGXQAcFvWSCQn2fuwDYZoZRk3oFjefr51WBNDGG9EjF1YDavg7pwGDFSAVWC5K42CBcLLv5U OR DONATE HERE: https://www.monerotalk.live/donate GUEST LINKS: https://www.wrapsynth.com/ TIMESTAMPS (00:00:00) Monerotopia Intro. (00:30:39) Monerotopia Price Report Segment w/ Bawdy. (01:30:57) Monerotopia News Segment w/ Doug. (01:33:18) Telegram to roll out native non custodial $GRAM wallet to over 1 billion users. (01:37:48) Kraken US will begin applying monthly transaction limits to Monero on August 15. (01:39:13) Zachxbt "all hardware wallets are complete garbage". (01:40:42) SenLummis post. (01:45:53) India Orders GitHub to Remove Bitchat Repositories. (01:47:29) Monero Research lab post. (01:59:55) Monerotopia Viewers on Stage Segment. (03:28:29) Monerotopia Final. NEWS SEGMENT LINKS: SPONSORS: PRICE REPORT: https://exolix.com/ GUEST SEGMENT: https://cakewallet.com & https://monero.com NEWS SEGMENT: https://www.wizardswap.io XMR.BAR: https://xmr.bar Don't forget to SUBSCRIBE! The more subscribers, the more we can help Monero grow! XMRtopia TELEGRAM: https://t.me/monerotopia XMRtopia MATRIX: https://matrix.to/#/%23monerotopia%3Amonero.social ODYSEE: https://bit.ly/3bMaFtE WEBSITE: monerotopia.com CONTACT: monerotopia@protonmail.com MASTADON: @Monerotopia@mastodon.social MONERO.TOWN https://monero.town/u/monerotopia Get Social with us: X: https://twitter.com/monerotopia INSTAGRAM: https://www.instagram.com/monerotopia DOUGLAS: https://twitter.com/douglastuman SUNITA: https://twitter.com/sunchakr TUX: https://twitter.com/tuxpizza

AABP Executive Director Dr. Fred Gingrich is joined by Dr. Travis White, Chief Veterinary Officer for Endovac Animal Health. Endovac is the sponsor of the student reception at the 59th AABP Annual Conference in Minneapolis, Minn. Find information about Endovac and their portfolio of animal health products on this page.  Gram-negative diseases remain a major challenge across beef, dairy, calf ranch and feedlot systems. White discusses these challenges and why the focus should not necessarily be on the organism itself, but thinking about endotoxin, immune stress and animal resilience. Endotoxin exposure drives the clinical signs associated with these disease challenges leading to inappetence, depression, fever and leukopenia.  White reviews core-antigen technology and how this approach differs from a conventional multi-antigen bacterin. He also discusses Immune Plus from Endovac and how veterinarians should think about immune stimulation. This technology can fit into existing vaccine and herd health programs vs. replacing currently used tools. White also reviews a challenge study with Salmonella Dublin and the take-aways from this study to help veterinarians who want to build stronger Gram-negative disease control programs.  

The Viksit Bharat Guarantee for Rozgar and Ajeevika Mission (Gramin) (or VB-G RAM G) has replaced the 21-year-old Mahatma Gandhi National Rural Employment Guarantee Act (or MGNREGA). The VB-G RAM G Act provides a statutory guarantee of 125 days of wage employment per financial year to rural households whose adult members volunteer for unskilled manual work (up from 100 days under MGNREGA). However, under the new system, the States' contributions to the scheme will increase by 300%. In this episode we discuss how this change reflects on Indian welfare policy, and more. Guest: Rajendran Narayanan, Associate Professor at Azim Premji University, Bengaluru Host: Nitika Francis Producer and editor: Jude Weston Learn more about your ad choices. Visit megaphone.fm/adchoices

Episode #290 with Adam and Taylor. Come send it with the boys, as we discuss - Being banned from the Gram, Obsession, Modern cinema, Hugh Jackman, Scumbag Lawyers, Prego Puck, 2nd degree murder, Getting sucked out, Ding dong ditching sex offenders, and much more... Follow us on Instagram & TikTok: https://www.instagram.com/bigsendpodcast https://www.tiktok.com/@bigsendpodcast Patreon BoSodes(Bonus Episodes): https://patreon.com/BigSendPodcast Please forward all complaints to: bigsendpodcast@gmail.com

We chat Kyle's sneaker obsession, dressing like a "normie", microwaving eggs, the smallest giggable amps possible, and more!   You can help support the show on our PATREON for as little as $1 a month! Double down to bump it to $2 a month and you'll get an extra episode every week!   Join the fun on our Facebook group!   Follow us on the fuckin' Gram!   Subscribe to our YouTube Channel for the video version of the show, demos, vlogs, and more!   We have shirts available at The Jerk Store!   Check out our band Plane Without a Pilot   Hosted by Brian Gower and Kyle McIntyre

En este episodio de gramática y lengua española vamos a revisar las suposiciones empleando el modo indicativo, especialmente los tiempos verbales del presente, futuro y condicional. 

Fredrik och Kristoffer sågs på stan för ett snack om Gram, teoribyggande, och ganska mycket mer. Kristoffer jobbar på ett jätteprojekt när han inte har något annat för sig. Kommunikation är ett problem - när det tar massor av tid att förklara varför och hur någonting behöver göras, tid man också skulle kunna lägga på att få en del av dem gjorda. Problemen kanske också behöver mogna och utforskas innan det går att låta någon annan göra något åt dem på ett effektivt sätt. Spelar det någon roll hur mycket ett gränssnitt sticker ut eller passar in i nuvarande trender? Vad får uppmärksamhet och varför? Och varför får det inte fler konsekvenser att påstå saker vitt och brett? Man borde inte bygga verktyg som stödjer dåliga sätt att jobba på. Man borde hitta bättre sätt att jobba istället för att bygga bättre verktyg för att stödja dåliga arbetssätt. Eller? Branschen har fastnat i icke-optimala sätt att jobba. Vad vill Kristoffer göra med Gram och varför? Och vad vill han inte göra? Hur borde flikar fungera? Och hur borde bra git-stöd i en textredigerare se ut? Varför är långa listor i gränssnitt alltid svårt? Det och andra problem man borde lösa på ett bra sätt en gång, inte halvbra många gånger. Programmering som teoribyggande. Varför kan vi inte rita i våra IDE:er, eller på andra sätt fånga allt teoribyggande och alla antaganden som inte går att utläsa av koden? Gränsen mellan språkmodeller och deterministiska verktyg. AI-sök saboterar möjligheter att hitta vissa sorters nålar i höstacken. Kodgranskning och teoribyggande - borde man fokusera mer på teorin i sitt granskande? All teori som inte finns med i koden. Rust är lite för bra för Kristoffer. Men det finnas massor som skulle kunna bli bättre också. Ett stort tack till Cloudnet som sponsrar vår VPS! Har du kommentarer, frågor eller tips? Vi är @kodsnack, @thieta, @krig, och @bjoreman på Mastodon, har en sida på Facebook och epostas på info@kodsnack.se om du vill skriva längre. Vi läser allt som skickas. Gillar du Kodsnack får du hemskt gärna recensera oss i iTunes! Du kan också stödja podden genom att ge oss en kaffe (eller två!) på Ko-fi, eller handla något i vår butik. Länkar Sommar-låten Hajk jj - git-ersättare Zed Patrik - vän av och gäst i podden Bug refinement - man vet att det är viktigt när förklaringen är femton sidor om man skulle skriva ut den! Mörk - Kristoffers markdownparser i Gleam Om jag haft mer tid hade jag skrivit ett kortare brev Gram Neovim Kai's power tools Bryce Kai's power goo - video Kai själv Jupyter notebook Magit Versionskontrollsystem där man skapade motsvarigheten till en commit först - innan man börjar göra ändringar - var nog just jj i ett avsnitt av Developer voices RAD debugger RAD game tools REPL - read-evaluate-print-loop Tailwind Immediate-mode GUI CSS GPUI - hårdvaruaccelererat UI-ramverk som driver Zed och Gram Gleam Commonmark Servo Nlnet - ger stöd till personer och organisationer som bidrar till ett öppet informationssamhälle Henna Virkkunen Computer science off course Programming as theory building Peter Naur BNF - Backus-Naur-form, eller Backusnormalform MS paint Stöd oss på Ko-fi! objc2 - Rust-låda med Objective-c-bindningar Newspeak Deltadb - Zeds nya lösning för versionskontroll Zig Tokio Garry Tan - VD för Y combinator och glad vibekodare Token ring Titlar Ett nytt utvecklingspass Den tekniska personen När vi har mer än tre användare Ticket-fokuserat sätt att arbeta Lär er använda git Jag borde bli bättre på git Alla borde bli bättre på git Atomerna för att bygga saker Ett sämre git Sortera sin kompost Välja precis rätt ord Gram power tools Metaboll Experimentgram Det är ju bara en texteditor En ny padda Immediate-mode CSS Tre tentakler Ni ser inte bra ut utifrån I ord förklara vad jag vill ha När man kan sin modell Vem är du att vara upprörd? Par-promptande Glorifierat undo En teori om hur man vill jobba Framtiden för programmering Min startup i Gleam Rust är för bra Nu sker det magi i bakgrunden Värdelös kunskap Professionell Rust Ett skal som ser ut att funka Mesh och b-post

Nejen bronzovou figurku, ale i další zajímavé nálezy u Tučína – jako třeba stříbrný římský denár – veřejnosti představí nová naučná stezka. Obec ji chce otevřít ještě v průběhu prázdnin.

Episode kicks of with parody news intro. Not our classiest. Then we've got more on the Gram Man's passing. We move onto if taco bell is causing this nations loose stools. I know I sat on one. Then we've got how Tarrif's are getting in the way of your kids ability to play boring board games and why that might actually keep them from becoming patsies in murder trials. Get your porch tickets: www.porchtour.comHave your porch porched: TheFireTix.comwear my merch on your porch: RobBernsteinMerch.comSit on your porch and watch my extended extremely premium episode: www.RobBernsteinComedy.comWear Sheath On A porch: Sheath.com promocode: RYM

Most people open Claude or ChatGPT for the first time, type a question, and start working. And that's fine,  it'll still give you a decent answer. But if you stop there, you're using maybe 20% of what either tool can actually do for your business. The other 80% lives in the setup: a handful of settings that tell the AI who you are, how you want it to sound, and what it's allowed to access on your behalf. Do this once, it takes about 15-20 minutes, and every conversation after that starts already knowing your business instead of starting from zero. Here's exactly what to set up, in both tools, before you dive in. Resources: The Meeting Place Membership Rock The Reels 1:1 Coaching Free Client Welcome Guide Additional Trainings and Resources Connect with Brittni: Follow me on the Gram - @brittni.schroeder Join my Facebook Group  Visit my website Subscribe to my Youtube You can find the complete show notes here: https://brittnischroeder.com/podcast/before-you-start-using-claude-and-chatgpt-the-setup-most-people-skip

"Everyone gets cancer five times a day — your body already knows how to take care of it." Sam Bertram is the co-founder and CEO of OnePointOne, the vertical-farming company behind the Willo brand. With his brother John (CTO), Sam is building fully automated indoor farms — powered by AutoStore robotics, AI, and precision LED light — that grow nutrient-dense produce in about 15 days, with no soil and no sunlight. But the vision goes far beyond salad: using plants as living factories to manufacture pharmaceutical-grade molecules (like monoclonal antibodies) at a fraction of today's cost, and eventually "closing the loop" on human nutrition — food grown to your own blood work. *Reduce your risk of Alzheimer's with my science-backed protocol for women 30+: * https://go.neuroathletics.com.au/yout... Subscribe to The Neuro Experience for evidence-based conversations at the intersection of brain science, longevity, and performance. _____ TOPICS DISCUSSED 00:00 – Cold open: bold claims & who is Sam Bertram 00:51 – Why vertical farms collapse (Bowery) & how OnePointOne survived 05:49 – The number that started it all: 1.1 billion malnourished 07:30 – Why micronutrient deficiency is the real killer 08:21 – What OnePointOne actually is: "a box that grows plants" 10:03 – Training plants with light & water to boost nutrients 11:00 – Closing the loop: food matched to your body's data 12:28 – The plan for 1,000 farms & the AutoStore partnership 15:46 – Plant biology 101: growing with no soil, no sun 18:11 – The robots — and why humans are still essential 21:27 – Pesticides, washing produce & disease clusters 23:53 – Beet greens for cancer patients & the Whole Foods deal 24:42 – A 13-person company run on AI 25:15 – The 15-day crop & cracking the cost code 27:11 – Biology vs. software: what really makes it work 28:01 – Humanoid robots, ~5 years away 30:27 – Why raise $80M and not $500M 33:56 – What they do to a plant that nature never would 36:28 – Expanding crops: strawberries, cannabis, mushrooms 38:05 – Growing food & materials in space (Elon, text back) 39:38 – Personalized food grown from your blood test 40:24 – Food as medicine & the truth about Western medicine 41:11 – Sam's mystery illness: 8 years without exercise 44:23 – Plants as medicine — who "fed and dosed" the world? 49:32 – Manufacturing pharmaceutical molecules inside plants 50:18 – Replicating $30–40k drugs cheaply (Rituximab, Lecanemab) 54:21 – Why the Middle East may beat the US, RFK Jr & MAHA 55:11 – Genetics, cancer & the healthy 80–90% 57:35 – Taste: beating junk food at its own game 58:27 – Why good food got so expensive (money printing) 59:14 – A world without hunger — how far away? 1:00:24 – Tasting the farm: microgreens, bok choy, mizuna & kale _______ Thank you to our sponsors Jones Road Beauty: https://jonesroadbeauty.com Use code LOUISA for a free full-size mascara with your purchase (and mention The Neuro Experience podcast when they ask where you heard about them) Qualia: https://www.qualialife.com/neuro Use code NEURO for 50% off Qualia Magnesium+ Cure Hydration: https://curehydration.com Use code NEURO for 20% off (also available on Amazon) _______ I'm Louisa Nicola - clinical neurophysiologist - Alzheimer's prevention specialist - founder of Neuro Athletics. My mission is to translate cutting-edge neuroscience into actionable strategies for cognitive longevity, peak performance, and brain disease prevention. If you're committed to optimizing your brain- reducing Alzheimer's risk - and staying mentally sharp for life, you're in the right place. Stay sharp. Stay informed. Join thousands who subscribe to the Neuro Athletics Newsletter → https://bit.ly/3ewI5P0 Instagram:   / louisanicola_   Twitter :   / louisanicola_   Learn more about your ad choices. Visit megaphone.fm/adchoices

We chat camming, Ai Jerks, Ai pedals, how long it takes to sell used gear nowadays and more!   You can help support the show on our PATREON for as little as $1 a month! Double down to bump it to $2 a month and you'll get an extra episode every week!   Join the fun on our Facebook group!   Follow us on the fuckin' Gram!   Subscribe to our YouTube Channel for the video version of the show, demos, vlogs, and more!   We have shirts available at The Jerk Store!   Check out our band Plane Without a Pilot   Hosted by Brian Gower and Kyle McIntyre

En este episodio de gramática y lengua española continuamos hablando de la diferente entre el condicional y el subjuntivo, pero, en este caso, de los tiempos compuestos. 

Producers for MMO #226 Executive Producers Berlin   Fiat Fun Coupon Producers Eli the Coffee Guy Dugitup Sharky Shark Emily the Fed Nail Lord of Gaylord Praetor Wiirdo of the Not So Flat Lands Preator Porrecca of Peoria Doiceses: Hempress Emily M.   Booster Producers boolysteedfountain.fm | 3,333 | BAG DADDY BOOSTER! NostrGangfountain.fm | 101   Creative Producers: Episode Artwork Woof!   Follow Us: X/Twitter MMO Show John Dan Youtube (while it lasts) MMO Show Livestream Rumble MMO Show Livestream Twitch MMO Show Livestream   Shownotes: Dan's Sources China's New Ethnicity Law Explained in 3 Minutes | APT Democrats call on Platner to drop out of Maine Senate race amid sexual assault allegation Behind the apparent rise of democratic socialism and what it could mean for U.S. politics National Guard soldiers with Memphis Safe Task Force fatally shoot 20-year-old man AfD Tops Polls as Germany's Coalition Faces Growing Pressure German prosecutors say Ukraine ordered 2022 Nord Stream pipeline sabotage | DW News Two explosions reported in Syria's Damascus during French president's visit Body found in search for woman accused of targeting Ukrainian tycoon in Monaco Passenger responds to Turkey blocking cruise ship with LGBTQ+ theme How Turkey became indispensable to NATO | DW News NATO Summit: What's at Stake for Erdogan? China's New Ethnicity Law Explained in 3 Minutes | APT Democrats call on Platner to drop out of Maine Senate race amid sexual assault allegation Behind the apparent rise of democratic socialism and what it could mean for U.S. politics   John's Shownotes   Patriot Front            Patriot Front March CNN            Patriot Front Manifesto   NYNY            Fmr Phizer Building Buckling   Media            Nina totenberg Alito Report   Syria            Macron in Syria Explosions CBS             Iran            FOX Iran Strikes Today   Africa            Somaliland Base Pan Africa Troy Beverly   Right to Fix            Right to Fix L1 Automotive   Parasite            Cyclosporiasis GMA            Illinois Local Report Cyclosporiasis   Entrepreneurship            New Maid Service Lubbock  

Instagram expert Sue B. Zimmerman joins the show to talk about her 30+ years as an entrepreneur, her 18 businesses, and how she built her career teaching Instagram marketing to women and small business owners. She and Brittani dig into why authenticity beats AI-generated content, how Sue B. is growing a new audience on Substack through her Soulful Selling Sisterhood community, and how a new skincare affiliate venture became an unexpected business highlight. The two also talk about the power of community-driven, grassroots marketing over chasing algorithms.   Resources: The Meeting Place Membership Rock The Reels 1:1 Coaching Free Client Welcome Guide Additional Trainings and Resources Connect with Brittni: Follow me on the Gram - @brittni.schroeder Join my Facebook Group  Visit my website Subscribe to my Youtube You can find the complete show notes here: https://brittnischroeder.com/podcast/#  

Andrew Walsh joins B-Gow to chat cats, stainless steel frets, staying out of the loop, what it takes to be a “real bass player”, starting vs. joining a band and more!   You can help support the show on our PATREON for as little as $1 a month! Double down to bump it to $2 a month and you'll get an extra episode every week!   Join the fun on our Facebook group!   Follow us on the fuckin' Gram!   Subscribe to our YouTube Channel for the video version of the show, demos, vlogs, and more!   We have shirts available at The Jerk Store!   Check out our band Plane Without a Pilot   Hosted by Brian Gower and Kyle McIntyre

This episode is all about how to survive your HSG. Otherwise known as a hysterosalpingogram. That's a big scary word so let's break it down! Hystero = uterus Salpingo = tube Gram = picture of Hysterosalpingogram (HSG) is basically a picture of your fallopian tubes. I like to think of the fallopian tube as the embryo transport system. It's where the egg and sperm come together and it's how the embryo will travel to the uterus. Part of fertility screening is not just for FSH, estradiol, and AMH for women. It's not just to see how fast the swimmers are swimming. It's also important to make sure that the fallopian tubes are open! It's the "T" in the tushymethod.com In today's episode of The Egg Whisperer Show, I'm talking more about HSG. Read the full show notes on Dr. Aimee's website Subscribe to my YouTube channel for more fertility tips!  Subscribe to the newsletter to get updates Dr. Aimee Eyvazzadeh is one of America's most well known fertility doctors. Her success rate at baby-making is what gives future parents hope when all hope is lost. She pioneered the TUSHY Method and BALLS Method to decrease your time to pregnancy. Learn more about the TUSHY Method and find a wealth of fertility resources at www.draimee.org.

Brandon Aiyuk is STILL making waves on social media, and there are other wide receivers out there too. We cover all angles of the Aiyuk situation, plus other Commanders news, in this week's Best of Commanders!

Brandon Aiyuk is STILL making waves on social media, and there are other wide receivers out there too. We cover all angles of the Aiyuk situation, plus other Commanders news, in this week's Best of Commanders!

Brandon Aiyuk is STILL making waves on social media, and there are other wide receivers out there too. We cover all angles of the Aiyuk situation, plus other Commanders news, in this week's Best of Commanders!

Brandon Aiyuk is STILL making waves on social media, and there are other wide receivers out there too. We cover all angles of the Aiyuk situation, plus other Commanders news, in this week's Best of Commanders!

Couldn't think of a funny or interesting title. Just getting back into the swing of things...    You can help support the show on our PATREON for as little as $1 a month! Double down to bump it to $2 a month and you'll get an extra episode every week!   Join the fun on our Facebook group!   Follow us on the fuckin' Gram!   Subscribe to our YouTube Channel for the video version of the show, demos, vlogs, and more!   We have shirts available at The Jerk Store!   Check out our band Plane Without a Pilot   Hosted by Brian Gower and Kyle McIntyre

En este episodio de gramática y lengua española continuamos hablando del condicional, pero añadiendo un punto complejo para algunos estudiantes: la diferencia con el pretérito imperfecto de subjuntivo. 

Try TrueDark glasses: https://truedark.comTry Danger Coffee: https://dangercoffee.com/Try Suppgrade Labs: https://shopsuppgradelabs.com/Over the last 15 years, I've built companies and spent millions testing what actually upgrades human performance, and I can tell you that mental fatigue has absolutely nothing to do with willpower. In this video, I break down the exact neurochemistry of deep focus. When your brain runs out of dopamine, norepinephrine, and acetylcholine, or gets bogged down by oxidative stress, your executive function collapses. The pharmaceutical industry wants you hooked on expensive stimulants that only treat the symptoms, but the real biological fix is a common antioxidant: Vitamin C.I explain how your brain hoards Vitamin C at levels 100 times higher than your blood, how to bypass the blood-brain barrier using SVCT2 transporters, and why eating sugar before you take it completely blocks its absorption. You'll also learn the exact biohacking stack—including NAC, CoQ10, PQQ, and L-theanine—you need to recycle this molecule, protect your mitochondria, and maintain effortless, razor-sharp focus all day long.Thank you to our sponsors!-KILLSwitch | If you're ready for the best sleep of your life, order now at https://www.switchsupplements.com/and use code DAVE for 20% off-Viome | Check it out at viome.com and use code 10DAVE for 10% off. It's time to stop guessing and start knowing your body.-iRestore | Reverse hair loss at www.irestore.com/DAVE and get exclusive savings on the iRestore Elite, use code DAVETimestamps:00:00 – Trailer00:21 – Focus Is Biology01:53 – Brain Chemistry Explained05:33 – Vitamin C Revealed07:09 – How to Dose08:18 – Absorption Barriers09:31 – Optimize Uptake10:00 – Oxidative Stress Problem11:00 – Stack for Deeper FocusConnect with Dave Asprey!Website: https://daveasprey.comTikTok: https://www.tiktok.com/@daveaspreyofficialInstagram: https://www.instagram.com/dave.asprey/Facebook: https://www.facebook.com/Daveaspreyofficial/X: https://x.com/daveaspreyYouTube: https://www.youtube.com/c/daveaspreybprThe Human Upgrade Podcast: https://www.instagram.com/TheHumanUpgradePodcast/ https://m.facebook.com/Thehumanupgrade/Danger Coffee: https://dangercoffee.com/DAVE15Dave Asprey's BEYOND Conference: https://beyondconference.com/Dave Asprey's New Book - Heavily Meditated: https://daveasprey.com/heavily-meditated/Dave's favorite supplements: https://www.shopsuppgradelabs.com/discount/DAVE15Upgrade Labs: https://upgradelabs.com40 Years of Zen: https://40yearsofzen.comSee Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

En este episodio de gramática y lengua española continuamos hablando de los tiempos verbales. Hoy es el turno de los condicionales. Hablamos del condicional simple y el condicional compuesto. 

No Agenda Episode 1874 - "Kennel Index" Kennel Index Executive Producers: Sir Loren Carl Nagel Catty Bones Walker Ostler Sir Cliffy — Helena Agri-Enterprises (Greg Clifton) David Chapman Dennis Cadle — Manuka Gold Sir Michael of the Midwest and Dr of Philosophy — 2pifi Starlink Associate Executive Producers: Eli the Coffee Guy Stefan Trockels Linda Lupatkin — Imagemakers Ink Bob from Monmouth OR Knights and Dames: Carl Nagel > Sir Crazy Carl of the Great Carolina Pine Forest Walker Ostler > Sir Zeppelin of the Snake River Plain Catty Bones > Red Dame (matching Sir J-Bones posthumous Red Knight) End of Show Mixes: Jus Baker (No Remorse / Pentagon Orbs Remix / Slide thru the Gram v1) MVP (Being Free135 / Cool Clear Data) Oystein Berge (Oh McElon Session) Art By: Francisco Scaramanga Mark van Dijk - Systems Master Ryan Bemrose - Program Director Back Office Jae Dvorak Chapters: Dreb Scott Clip Custodian: Neal Jones Clip Collectors: Steve Jones & Dave Ackerman ShowNotes Archive 1867.noagendanotes.com No Agenda Peerage RSS Podcast Feed Last Modified 06/04/2026 16:26:36 by Freedom Controller