Podcasts about continuous

Dr. Mohammad Ashori is a board-certified family medicine physician, health coach, and rock climber. We talked about the importance of putting on muscle, how much rest we need between training sessions, nutrition and metabolic health, zone 2 cardio, key metrics and blood markers to track, different levels of data, why some data can be misleading or harmful, relative vs. absolute risk, and much more. Want to work with Dr. Ashori? Get 15% off your 1st health assessment with Dr. Ashori. Book your schedule here and use the code: NUGGET15 Want climbing or nutrition coaching? Free Masterclass (Learn more about Climb Accelerator)

Relax with four hours of continuous relaxing stories for sleep in this Places of Wonder sleep story collection for grown-ups. Enjoy a captivating selection of calm bedtime stories for grown-ups, narrated by a soothing female voice, with each story transporting you to an extraordinary destination, from majestic fjords and serene lakeside retreats to grand museums, regal castles, an ice castle, and a beautiful lakeside villa in Italy. Created for bedtime relaxation and sleep, these soothing stories offer a peaceful mental escape while you drift toward sleep. Listen without interruption and use the chapters to customize your experience. It's time to dream away.Original Script, Narration, Calm Music, Video, Production, and Sound Design by Michelle Hotaling, Dreamaway Visions LLC 2026 All Rights Reserved✨YOUTUBE:

Peter Huessy argues that nuclear deterrence remains vital but requires continuous modernization of missiles, submarines, bombers, and command systems to match Russian and Chinese advancements. He explains that conventional defense fails against nuclear attacks in military war games. Huessy advocates for "counterforce" strategies—targeting an enemy's weapons and leadership rather than innocent civilian cities ("countervalue"). He also supports adding theater nuclear systems to deter Russia's "escalate to win" doctrine, which utilizes low-yield battlefield nuclear weapons to force a NATO stand-down. (2)

Kim Wiggins is a world-renowned painter from New Mexico, and one of the most recognizable artists working in the American West today. His bold, swirling canvases have made him one of the leading artistic voices of what's come to be called the New West. His work hangs in the Autry, the Booth, and the American Museum of Western Art in Denver, and this past year he won Best in Show at the Prix de West, the most prestigious award in Western art, becoming the first modernist ever to take it home. Kim comes by all of this honestly. Both sides of his family came west in covered wagons in the late 1800s and settled into ranching along the Pecos River. His mother was running two ranches and a nightclub in Arizona by her mid-twenties; his father was a photojournalist for Life and Sports Illustrated, and a bush pilot who once had to sprint back to his plane in the Amazon and count the arrows in the side of the fuselage. Kim grew up surrounded by creativity– Thomas Hart Benton paged through his sketchbook and he ate lunch with Georgia O'Keeffe at Ghost Ranch. He began focusing seriously on painting while stationed in Europe with the military and soon became the youngest member of the Society of American Impressionists — a career most artists would give an arm for, and one he walked away from completely. We talk about why he walked away, which starts with a moment at a show in St. Louis when he spotted one of his paintings across the room, hurried over, and discovered it wasn't his. We cover the lean years that followed, including knocking on doors in Lubbock in the summer heat with a car full of paintings. And we get into the ideas behind the work: why a loss of innovation is the death knell of a civilization, why he watches children in galleries instead of collectors, the paintings he made for the NAACP after his son asked him a question he couldn't answer, and his conviction that a gift, by definition, belongs to someone else. I first met Kim back in 2022 at Maxwell Alexander Gallery's ten-year show, where he was on a panel I moderated, and I've wanted a longer conversation ever since. Hope you enjoy. --- Kim Wiggins Prix de West Maxwell Alexander panel discussion (2022) Full episode notes and links: https://mountainandprairie.com/kim-wiggins --- THANK YOU TO OUR SPONSORS: Mountain & Prairie is listener supported via Patreon, and brought to you with support from the Colorado Cattlemen's Agricultural Land Trust, MIRASOL: Looking at the Sun, Patagonia Books, the Rye Resurgence Project for their generous sponsorship. --- TOPICS DISCUSSED: 0:00 - Introducing Kim Wiggins 3:44 - Wiggins' family history 9:49 - Kim's dad's photojournalism travels 12:44 - Parental pressure? 17:07 - Military education 20:36 - Money v. creativity 28:36 - Making the decision to change course 34:52 - When things started to work  38:42 - It's never about Kim 43:47 - Continuous evolution 49:44 - How Kim's paintings come together 53:10 - Inspiration from his son 58:45 - Collaborative projects 1:04:26 - Blood and Thunder 1:06:21 - All the circles aligning 1:11:03 - How art has changed in the West 1:14:00 - Book recs 1:17:47 - Wrapping up --- ABOUT MOUNTAIN & PRAIRIE: Mountain & Prairie - All Episodes Mountain & Prairie Shop Mountain & Prairie on Instagram Upcoming Events About Ed Roberson Leave a Review on Apple Podcasts

In this episode of The Agent of Wealth Podcast, host Marc Bautis is joined by Michael Maloney, is joined by Michael Maloney, Founder and Chief Creative of Brand Force 5, to explore why branding is about much more than a logo — and how the details many business owners overlook can shape customer trust, perception and ultimately, growth.From the quality of a postcard to the appearance of a service vehicle, every interaction can either strengthen or weaken a brand. Marc and Michael discuss how businesses can create a consistent identity, avoid costly branding mistakes and think of branding as an investment rather than simply another expense.In this episode, you will learn:Why a logo is something people see, but a brand is something people feel and experience.The five principles behind the FORCE framework — Focused, Ownable, Relevant, Continuous and Evocative — and how they can guide every aspect of a brand.Why inconsistent messaging, visuals and customer experiences can quietly erode trust and credibility.How to think about the hidden cost of bad branding — including the customers and opportunities you may never realize you're losing.What new business owners should define and invest in before spending money on marketing materials and campaigns.Why AI can be a valuable branding tool, but may still struggle to replace human judgment, emotion and creative insight.And more!Tune in for a conversation about the small details that influence how customers perceive your business and why building a strong brand starts long before you choose a logo.Resources:Episode Transcript & Blog | brandforce5.com: (862) 212-3720 | Brand Force 5 on Facebook | Brand Force 5 on LinkedIn | Bautis Financial: 8 Hillside Ave, Suite LL1 Montclair, New Jersey 07042 (862) 205-5000 | Schedule an Introductory Call

Gentle brown noise provides soothing relief for tinnitus and promotes a calm, balanced mental state.https://distrokid.com/hyperfollow/brownnoisesleepsounds/brown-noise-sound-3Buy me a Coffee Support me here ☕ Thank you!buymeacoffee.com/BrownNoiseSleepSounds

Bienvenido a la cuarta y última lección de nuestra serie sobre el presente perfecto continuo. Daniel, ¿qué vamos a hacer hoy exactamente? Hoy es diferente a las otras lecciones. En lugar de explicar reglas nuevas, vamos a escuchar la historia de un inmigrante que lleva un año en Canadá. También vamos a ver el presente perfecto continuo en acción en una conversación real y natural. Después del diálogo, vamos a analizar juntos los momentos más importantes. Es una lección de repaso, pero contada como una historia. Presta atención al presente perfecto continuo y quédate hasta el final, porque vas a repasar todo lo que aprendiste en esta serie. ¡Empecemos! Recuerda que todos los recursos para este episodio, incluyendo la transcripción, la tabla de vocabulario y ejercicios para repasar el aprendizaje, están disponibles en nuestro sitio web.  Haz clic en este enlace para ver todos los recursos para este episodio:   https://inglesdesdecero.ca/275 ----- Dale “me gusta” a nuestra página en Facebook: https://www.facebook.com/inglesdesde0/ ----- Síguenos en Instagram: https://www.instagram.com/ingles.desde.cero/ ----- Suscríbete en YouTube: https://www.youtube.com/@inglesdesdecero145 ----- Encuéntranos en Pinterest: https://es.pinterest.com/inglesdesdeceroca/ ----- Aprende inglés con nativos que se formaron en su enseñanza. ¡Visita nuestro sitio web, https://inglesdesdecero.ca/ para inscribirte y seguir todas nuestras lecciones! No dejes pasar esta oportunidad con Shopify y regístrate para un período de prueba por solo un dólar al mes en shopify.mx/desdecero Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Most people with type 2 diabetes are told it's a forever sentence. Chris Reade was one of them. In 2017, his A1C hit 9.1 with zero warning. He refused medication, did the research, and built a food system around the science of soluble fiber. Six months later, his A1C was 5.5. Seven years later, it has never gone back up. In this episode, entrepreneur and author Chris Reade explains how he reversed a shocking diabetes diagnosis using soluble fiber instead of medication, and why simple, sustainable habits beat extreme diets every time.   RESOURCES: Learn more about Chris Reade here: https://www.beatingdiabetes.us/ Instagram: @beatingdiabetesus Get 10% off Peluva minimalist shoe with coupon code COACHTARA here: http://peluva.com/coachtara   CHAPTERS: 00:00 Intro: meet Chris Reade, author of Beating Diabetes 01:13 Sponsor: Peluva Shoes ad 02:59 Chris's diabetes story 03:35 Insurance renewal reveals an A1C of 9.3 06:00 Doctor's grim prognosis and hunting for the cause 09:07 Cancer scare leads to a surprisingly simple fix 11:13 The research: soluble fiber studies since the 1970s 12:19 How soluble fiber controls sugar, weight, and health 17:52 Writing the book: family history and friend Tim's scare 19:41 Frustration with "pills-only" medicine 22:03 Habit stacking, forgiveness, and small daily habits 28:04 Segment: Tara's coaching, app, and retreats 33:00 Soluble vs. insoluble fiber, explained simply 36:34 Continuous glucose monitors: benefits and pitfalls 43:46 His current routine, and "the warranty ends at 30" 44:49 Soapbox: the vegan/carb-fear myth 51:27 Morocco motorcycle trip: eating simply on the road 54:58 Success story: Tim's A1C drops from 13 to normal 56:57 Next-level challenge: cooked-and-cooled resistant starch   WORK WITH TARA: Are You Looking for Help on Your Wellness Journey? Here's how Tara can help you: TRY TARA'S APP FOR FREE: http://taragarrison.com/app INDIVIDUAL ONLINE COACHING: https://www.taragarrison.com/work-with-me CHECK OUT HIGHER RETREATS: https://www.taragarrison.com/retreats   SOCIAL MEDIA:  Instagram @coachtaragarrison TikTok @coachtaragarrison Facebook @coachtaragarrison Pinterest @coachtaragarrison   INSIDE OUT HEALTH PODCAST SPECIAL OFFERS: ☑️ Upgraded Formulas Hair Test Kit Special Offer: https://bit.ly/3YdMn4Z ☑️ Upgraded Formulas - Get 15% OFF Everything with Coupon Code INSIDEOUT15: https://upgradedformulas.com/INSIDEOUT15 ☑️ Rep Provisions: Vote for the future of food with your dollar! And enjoy a 15% discount while you're at it with Coupon Code COACHTARA: https://bit.ly/3dD4ZSv   If you loved this episode, please leave a review! Here's how to do it on Apple Podcasts: Go to Inside Out Health Podcast page: https://podcasts.apple.com/us/podcast/inside-out-health-with-coach-tara-garrison/id1468368093 Scroll down to the 'Ratings & Reviews' section. Tap 'Write a Review' (you may be prompted to log in with your Apple ID). Thank you!

Finding great people is one of the biggest challenges remodelers face, but the answer isn't posting another job ad and hoping the right person comes along. In this episode, Kyle is joined by Justin Landry of JL Talent Solutions and Peer Group Coach Bryan Sebring for a practical conversation about creating a more intentional hiring process. Bryan shares how his approach to hiring has changed after working with Justin and why slowing down has helped him avoid costly hiring mistakes. They dig into what a strong recruiting process actually looks like, from defining the role before you start searching to using assessments and structured interviews to better understand each candidate. Justin also explains why the right people are out there, even when it feels impossible to find them, and how looking beyond the usual job boards can open up a much stronger candidate pool.If hiring has felt rushed, frustrating, or inconsistent in your remodeling business, this episode will help you think differently about how you find people and set them up to succeed!Running a remodeling company can feel lonely, but you don't have to figure everything out on your own.The Remodelers On The Rise Peer Groups connect you with non-competing remodelers who understand the challenges you're facing and are committed to helping each other grow.Learn more at  https://remodelersontherise.com/live-training-events-for-remodelers/remodelers-peer-group/Explore the vast array of tools, training courses, a podcast, and a supportive community of over 2,000 remodelers. Visit Remodelersontherise.com today and take your remodeling business to new heights!Key Takeaways Building clear role definitionsImplementing a systematic hiring processUsing assessments like DISC and previewCreating a strong onboarding experienceFostering a positive company cultureLeadership and conflict resolution in teamsThe importance of patience in hiringUsing data and science in recruitmentOnboarding best practices for successChapters00:00 Introduction and episode overview01:10 Start with the end in mind in hiring02:07 Justin Landry introduces JL Talent Solutions04:46 Brian Sebring shares his background and path to Justin08:25 The importance of patience in hiring10:15 Justin explains the vetting process and assessments13:25 Proactive vs reactive hiring strategies16:34 Best practices for posting and attracting candidates18:47 Using data science and assessments to select candidates22:42 Effective interview questions and techniques30:57 The significance of clear role definitions35:58 Onboarding best practices and 30-60-90 plan40:56 Leadership, culture, and team integration42:46 Continuous improvement and learning in hiring

"How long have you been living in Canada?" “Why have you been avoiding going outside?” Si entendiste esas preguntas, ya estás usando el presente perfecto continuo con WH questions sin darte cuenta. Y si no las entendiste del todo, en pocos minutos vas a poder hacerlo y responderlas con confianza. En esta tercera lección de la serie, vamos a combinar las WH questions — esas que se formulan con what, why, where, who, how y how long — con el presente perfecto continuo. Y lo vamos a practicar con un tema que todo inmigrante en Canadá conoce muy bien: el clima y la adaptación al frío. ¡Empecemos! Recuerda que todos los recursos para este episodio, incluyendo la transcripción, la tabla de vocabulario y ejercicios para repasar el aprendizaje, están disponibles en nuestro sitio web.  Haz clic en este enlace para ver todos los recursos para este episodio:   https://inglesdesdecero.ca/274 ----- Dale “me gusta” a nuestra página en Facebook: https://www.facebook.com/inglesdesde0/ ----- Síguenos en Instagram: https://www.instagram.com/ingles.desde.cero/ ----- Suscríbete en YouTube: https://www.youtube.com/@inglesdesdecero145 ----- Encuéntranos en Pinterest: https://es.pinterest.com/inglesdesdeceroca/ ----- Aprende inglés con nativos que se formaron en su enseñanza. ¡Visita nuestro sitio web, https://inglesdesdecero.ca/ para inscribirte y seguir todas nuestras lecciones! No dejes pasar esta oportunidad con Shopify y regístrate para un período de prueba por solo un dólar al mes en shopify.mx/desdecero Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

The fact that OpenAI has quickly adopted Temporal, a rapidly expanding AI ecosystem, and has even entered into a partnership with Crystal Palace FC shows that its business strategy is to pursue community-first innovation.For years, tech enterprises competed against each other on the grounds of innovative features. However, the competition has changed recently. Now the winners are those who build communities.In the recent episode of the Tech Transformed podcast, host Christina Stathopoulos, Founder of Dare to Data, is joined by Melissa Herrera, Senior Developer Advocate at Temporal, and Les Jackson, Staff Developer Advocate, to discuss the pivotal role of the community in technology. They further explore how Temporal's open-source philosophy fosters developer engagement, the impact of community feedback on product development, and the significance of partnerships, such as with OpenAI and Crystal Palace. The conversation emphasises the importance of authentic community relationships and the future direction of Temporal, highlighting the need for continuous integration and collaboration with developers.TakeawaysCommunity is a central part of Temporal's growth.Temporal's philosophy is rooted in open-source software.In-person community interactions are invaluable for developers.Feedback from the community directly shapes product direction.OpenAI's adoption of Temporal led to significant scaling.Partnerships should focus on community engagement, not just transactions.Temporal's collaboration with Crystal Palace merges tech and sports communities.Investing in community fosters trust and collaboration.Continuous integration with developer tools is essential for success.Authentic community relationships drive technology innovation.Chapters00:00 Introduction to Tech Transformed Podcast02:45 The Importance of Community in Technology05:13 Feedback from Developers: Shaping Product Direction07:51 OpenAI's Adoption of Temporal: A Case Study10:02 Unique Partnerships: Temporal and Crystal Palace15:12 Looking Ahead: Future of Temporal and Community Engagement19:38 Final Thoughts: Investing in CommunityTemporal, Enterprise AI, Developer Communities, Developer Advocacy, Open Source, OpenAI, AI Agents, AI Infrastructure, Durable Execution, AI Orchestration, Enterprise Software, Developer Experience, AI Workflows, AI Adoption, Community Led Growth, Developer Led Growth, Temporal SDK, Software Engineering, AI Engineering, Enterprise Technology, Crystal Palace, Tech Transformed

Listen to the full show podcast of The Continuous Call Team.See omnystudio.com/listener for privacy information.

Listen to the full show podcast of The Continuous Call Team.See omnystudio.com/listener for privacy information.

This week on Pulse: Hot Topics, Louise and George explore the rapidly shifting boundaries of science, AI and human health.They unpack the extraordinary research using AI to design entirely new, functioning viruses, and the biosecurity questions that follow. Then, turning the discussion to South Australia's newly announced Royal Commission into AI and whether governance can possibly keep pace with technological change.Then they look at the emerging era of the “continuous human”: Germany's €40 million push to crack continuous hormone monitoring, and Google's enormous SensorFM foundation model, trained on more than a trillion minutes of wearable data, which could point towards a future of personalised, continuous health intelligence.Plus, George recommends Paper Guide for scientific research, and we highlight two international opportunities for clinicians to contribute to research on AI in healthcare.Research Opportunities:Clinicians – be part of the research into clinician use of AI, open globally LinkAI policy vs frontline reality survey (deadline 2 September) LinkResources:SPRIND Continuous Hormone Monitoring Challenge LinkIda Tin Coined Femtech. Now She's Funding Continuous Hormone Monitoring, Forbes LinkSensorFM: Towards a general intelligence and interface for wearable health data, Google Research LinkRecommendations:Paper Guide LinkVisit Pulse+IT.news to subscribe to breaking digital news, weekly newsletters and a rich treasure trove of archival material. People in the know, get their news from Pulse+IT – Your leading voice in digital health news.Follow us on LinkedIn Louise | George | Pulse+ITFollow us on BlueSky Louise | George | Pulse+ITSend us your questions pulsepod@pulseit.newsProduction by Octopod Productions | Ivan Juric

In this episode, Matty A. and Ryan Breedwell analyze the most significant economic events shaping the market, from the possibility of an Iran peace deal to corporate America's record-breaking S&P 500 earnings beat. They examine the persistent housing affordability crisis, exploring how 30-year mortgage rates and changing generational habits are impacting homeownership.The discussion dives deep into the latest tech and private equity movements, including Morgan Stanley's $600 price target for SpaceX and Berkshire Hathaway's massive pivot into Google and home builders. With inflation data looming and job numbers facing continuous downward revisions, this episode provides critical insights for navigating today's complex investing environment.KEY TOPICS DISCUSSEDIran conflict peace negotiations and potential stock market reactionsCPI and PPI inflation data expectations and Federal Reserve rate policiesSpaceX market valuation and Morgan Stanley's $600 bull case price targetCoreWeave earnings reports and upcoming technical resistance levelsCorporate America's unprecedented 29.2% aggregate S&P 500 earnings beatUS housing affordability crisis and increasing 30-year fixed mortgage ratesBerkshire Hathaway deploying cash into Alphabet and Taylor Morrison HomePrivate equity firms holding 33,575 unsold businesses amid high borrowing costsKEY TAKEAWAYSHistorically high S&P 500 earnings beats indicate corporate margins are much stronger than Wall Street analysts anticipated.Continuous downward revisions in the US jobs report suggest ongoing economic cooling, which may take future Federal Reserve rate hikes off the table.High interest rates remain the crucial linchpin suppressing housing supply and affordability, leaving Gen Z increasingly sidelined from the American dream.Berkshire Hathaway's recent $6.8 billion investment in a home builder signals institutional confidence in the long-term necessity of new housing construction.Massive private equity portfolios backed by private credit are facing severe liquidity challenges as borrowing costs remain elevated.CONNECT & TAKE ACTIONImagos Income Fund: Text "INCOME" or "DEALS" to 844-447-1555 to learn more about Matty A's private debt fund targeting 10% fixed returns paid out monthly.Visit skylineocresidences.com to discover luxury condo ownership at Skyline OC, Orange County's tallest residential tower. Get a free financial audit on your investment portfolio by texting X-Ray to 844-447-1555

Sonia Garcia, Co-Founder and Chief Growth Officer at Amae Health, an integrated care company specializing in the treatment of severe mental illness that operates community-oriented clinics to deliver comprehensive psychiatry-led care. They provide continuity of care across multiple acuity levels, recognizing these conditions require long-term management. Amae is also advancing psychiatric care through research and partnerships by deploying wearables for health monitoring and investigating biomarkers and pharmacogenomics to improve treatment outcomes. Sonia explains, "Amae Health is an integrated care company that exists to treat the most complex mentally ill. So those that are suffering from schizophrenia, bipolar disorder, major depressive disorder. And we do so through these community-oriented clinics that deliver comprehensive care that's psychiatry-led, but also incorporates therapy, primary care, social services, peer support specialists, dieticians for a lot of nutrition guidance, and health coaching. Then clinic care coordinators manage all the in-between things that need to happen to get somebody back on track and living a healthy, stable life."   "Another core pillar within the company that I'll just add is getting into technical research and innovation, with lots of work we can do to uncover what is causing this, what interventions we can deploy. How can we partner with others in the ecosystem, whether that's pharma or life sciences, to advance psychiatric care and see if we can get towards putting many of these conditions into remission?" #AmaeHealth #MentalHealthCare #SevereMentalIllness #IntegratedCare #MetabolicPsychiatry #Schizophrenia #BipolarDisorder #ValueBasedCare #ContinuityOfCare #HealthcareInnovation #Psychiatry amaehealth.com Listen to the podcast here

Sonia Garcia, Co-Founder and Chief Growth Officer at Amae Health, an integrated care company specializing in the treatment of severe mental illness that operates community-oriented clinics to deliver comprehensive psychiatry-led care. They provide continuity of care across multiple acuity levels, recognizing these conditions require long-term management. Amae is also advancing psychiatric care through research and partnerships by deploying wearables for health monitoring and investigating biomarkers and pharmacogenomics to improve treatment outcomes. Sonia explains, "Amae Health is an integrated care company that exists to treat the most complex mentally ill. So those that are suffering from schizophrenia, bipolar disorder, major depressive disorder. And we do so through these community-oriented clinics that deliver comprehensive care that's psychiatry-led, but also incorporates therapy, primary care, social services, peer support specialists, dieticians for a lot of nutrition guidance, and health coaching. Then clinic care coordinators manage all the in-between things that need to happen to get somebody back on track and living a healthy, stable life."   "Another core pillar within the company that I'll just add is getting into technical research and innovation, with lots of work we can do to uncover what is causing this, what interventions we can deploy. How can we partner with others in the ecosystem, whether that's pharma or life sciences, to advance psychiatric care and see if we can get towards putting many of these conditions into remission?" #AmaeHealth #MentalHealthCare #SevereMentalIllness #IntegratedCare #MetabolicPsychiatry #Schizophrenia #BipolarDisorder #ValueBasedCare #ContinuityOfCare #HealthcareInnovation #Psychiatry amaehealth.com Download the transcript here

Experience gentle continuous rainfall sounds that create a peaceful ambient atmosphere perfect for deep sleep, meditation, and relaxation. Let the soothing rain wash away stress and help you find calm focus or restful sleep naturally.

In this episode of Maximize Your Hunt, host Jon Teater introduces a series focused on food plot seed choices, featuring insights from industry experts. The conversation emphasizes the importance of consulting in hunting property management, the principles behind creating effective seed blends, and the journey of various seed companies in the hunting industry. Listeners gain valuable knowledge on habitat improvement, soil health, and deer attraction strategies. In this conversation, the speakers delve into the principles of creating effective food plots, focusing on seed selection, genetics, and the importance of weatherproofing plots. They discuss the unique methodologies employed by different companies, particularly emphasizing the need for comprehensive coverage and the role of genetics in enhancing forage quality. The conversation also highlights the significance of feedback and testing in developing successful seed blends, showcasing a commitment to quality and effectiveness in wildlife management practices. Takeaways This podcast explores land management and hunting strategies. Consulting leads to greater success in hunting property management. Food plots are essential for designing a hunting property. Seed blends should focus on objectives like nutrient cycling. Diversity in food plots attracts deer effectively. Soil health is crucial for successful food plots. Understanding the home of the animal is key to habitat improvement. Testing and feedback are vital for developing seed blends. The journey of creating a seed company can be fulfilling. Continuous improvement is necessary in food plotting. I look at things differently than most people. We want to have complete coverage. You want to weatherproof your plots as much as you can. Healthy plants, healthy soil, healthy forages, healthy deer. We want to hear everything, the good, the bad and the ugly. The deer have to hit it, then they have to hit it hard. It's not complicated, it's not rocket science. We want as much tonnage as we can per acre. The industry focuses on a lot of fall focus. Ninety-five percent of what we test never makes it.   Tags hunting, land management, food plots, seed blends, deer attraction, ecological services, habitat improvement, consulting, whitetail deer, soil health food plots, seed selection, genetics, wildlife management, deer hunting, agricultural practices, sustainable farming, crop management, food plot strategies, land stewardship   Social Links https://vitalizeseed.com https://grandparayoutdoors.com/ https://www.realworldwildlifeproducts.com/ https://www.northwoodswhitetails.com/ https://whitetaillandscapes.com/ https://www.facebook.com/whitetaillandscapes/ https://www.instagram.com/whitetail_landscapes/?hl=en Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

I interviewed Jonathan Astruc about La Magie Opera on Friday, August 29, 2025 at Venice Immersive in Venice, Italy. Here is the story synopsis for La Magie Opera: "The viewer is part of a group of extras at the Palais Garnier. At the foot of the grand staircase, he crosses paths with Céleste, a young opera singer about to flee the stage. Nervous and unsure, she confides her doubts and dreams to him — torn between the weight of tradition and the desire to find her own voice. Freshly arrived from the provinces, this rising star is about to make her debut in a leading role, but her confidence is shaken. Drawn into her story, the user follows Céleste through the hidden corridors and majestic halls of the Opéra Garnier. As the building comes alive around her, she is transported into iconic scenes from legendary operas — each one echoing a part of her inner journey." Here are the contextual domains that are explored: Opera singing [5], but in the context of a long distance journey / adventure [9] as you explore lots of different spatial contexts from the surreal ocean bottom [9] to the mundane virtual castles [4]. Here is the Elemental Center of Gravity: 1st Center of Gravity of Earth Element / Environmental and Embodied Presence: Like a One-Shot Spatial Journey where there's no cuts, no edits, no fade-to-black, but rather a continuous experience where you seamlessly go from one scene to the next. Starting in the opera house, then going under underwater, and then walking through various castle-like scenes and environments.2nd Center of Gravity of Water Element / Emotional Presence: Music plays throughout, and there's a rhythm of the piece where each spatial context has a cut scene that plays out, and then it's all about moving bodies through space to get to the next song, piece of theatrically state acting + musical number, and by the end it feels like quite an awe-inspiring type of journey3rd Center of Gravity of Air Element / Mental and Social Presence: Multiplayer and social experience. There are other people in the experience, and sometimes they can deliberately clip into walls to test the bounds of the systsem, which impacts the plausibility of the virtually-mediated experience for others. Also it's very much a musical operal experience, but with actors playing out scenes. Can't remember if there's any dialogue or if it was all contained in the songs.4th Center of Gravity of Fire Element / Active Presence: There's some degree of freedom for moving through this experience, but very constrained and limited. It's a group experience, and so you're bound by other people, but very musch in the real of constrained agency for how you move through the space, but no narrative agency. They may have also had a few interactable moments with objects in the scene. But nothing that would dramatically shift the experience for different people. Archetypal Themes and Character Explored: Themes: "exploration of the ties between doubt, memory and the stage" -- Doubt, Illusion, and Memories Artist Statement: "This project is a sensory exploration of the ties between doubt, memory and the stage. More than a tribute to opera, it offers a vivid, inner reimagining where users become Céleste's silent partners. They journey through shifting worlds where music, design and storytelling blend into an intimate yet spectacular experience. Each scene is a visual metaphor for Céleste's emotions. The Palais Garnier transforms — from real stage to dreamlike realm — with mirrors and doubles revealing her turmoil and awakening. The viewer quietly follows her through reflections of passion, fear and longing. It's a poetic immersion into the artist's soul, merging opera's lyricism with immersive technology for an introspective, sensory voyage." https://www.youtube.com/watch?v=D_dnnx1ysHg This is a listener-supported podcast through the Voices of VR Patreon. Music: Fatality

When blue chip institutions suffer high-profile quality failures, the immediate corporate reaction is often to point fingers at the underlying technology. However, when an organization gets caught publishing fake frameworks and hallucinated citations for years, that isn't an AI failure. It's a systemic breakdown driven by an obsession with speed, profits, and volume over everything else.  This week, I unpack the recent PwC hallucination scandal reported by the Financial Times. While headlines focus on the embarrassing blunders, the real story is how closely this mirrors the pressure cooker inside many major companies. Countless organizations are making the mistake of rushing to automate, trading validity, brand equity, and human trust just to say they're doing "more, faster."  My goal this week is to help you avoid this digital speed trap by describing what it looks like to run a sustainable, human-centric framework for execution, quality, and accountability:​Exposing the Production Obsession Trap: Trading validity and accuracy for sheer velocity is a recipe for brand destruction. When you execute without grounding your outputs in human reality, speed simply means running off potential cliffs at higher velocity.  ​Closing the Human Reality Deficit: True leadership requires moving past vanity metrics like software logins or token counts. To eliminate organizational friction, you must capture hard data on human sentiment, behavioral fluency, and actual lived experience.  ​Executing a Continuous, Outcome-Driven Blueprint: Rather than treating strategic alignment as a static annual event, leaders need an agile, closed-loop system. By focusing on what matters most, diagnosing behavioral capability gaps, and continuously pulsing along the way, you protect credibility while driving real impact.  By the end, my hope is that you'll resist the urge to chase empty speed metrics or publish unverified "slop." Sustainable success in the AI age isn't about moving as fast as possible; it's about anchoring your technology in human truth, raising standards, and building lasting credibility.  —Share your honest thoughts on the impact of AI at work: https://howdopeoplefeel.comAnd if you'd benefit from help balancing performance, technology, and people, check out my website at ⁠https://christopherlind.co⁠—Chapters⁠00:00⁠ – The PwC Scandal: An Organizational Speed Breakdown⁠03:45⁠ – The Digital Speed Trap: Trading Validity for Velocity⁠07:30⁠ – Step 1: Focusing on What Matters Most (Pathfinder Pulse)⁠11:15⁠ – Step 2: Diagnosing Data Gaps & Human Friction (AER & HDPRF)⁠16:00⁠ – Step 3: Executing, Upskilling, and Continuous Pulsing (Helix)⁠20:10⁠ – Navigating Enterprise Bureaucracy & Continuous Agility⁠22:45⁠ – Three Surgical Leadership Moves to Reclaim Credibility  #Leadership #AIStrategy #FutureFocused #WorkforceTransformation #OrganizationalTrust

Send us Fan MailBlood sugar is everywhere right now. Glucose spikes, continuous glucose monitors, apple cider vinegar before meals, walking after meals, berberine, cutting carbs… but how much of this actually matters?In this episode, we're joined by Dr. Adrian Chavez, who has a PhD focused on insulin resistance and metabolic health, to break down what you actually need to know about blood sugar.We talk about what insulin resistance really is, why it can develop years before diabetes, whether fasting insulin and HOMA-IR are worth testing, what a normal blood sugar response after eating should look like, and whether healthy people actually need to worry about glucose spikes.We also get into CGMs, berberine, apple cider vinegar, post-meal walks, calorie intake, body fat, strength training, cardio, fibre, protein, and the lifestyle changes that actually make the biggest difference for metabolic health.Plus, we zoom out into a much bigger conversation about the wellness industry, health optimization, and why so many people are spending enormous amounts of time and money worrying about tiny details while overlooking the habits that have the strongest evidence behind them.If you've ever wondered whether you're insulin resistant, what blood work actually matters, or what you should be doing to protect your metabolic health long term, this episode will give you a much clearer place to start.In this episode, we discuss:• What insulin resistance actually is• How insulin resistance can develop before diabetes• Fasting insulin and HOMA-IR• Whether blood sugar spikes are actually harmful• Continuous glucose monitors for healthy people• Berberine and whether it actually works• Apple cider vinegar and post-meal walks• Strength training, cardio and daily movement• Calorie intake, body fat and metabolic health• Protein, fibre and blood sugar control• The problem with over-optimizing your health• What actually matters for long-term wellnessFollow Dr. Adrian Chavez on Instagram: @dr.adrianchavezListen to The Nutrition Science Podcast wherever you get your podcasts.Use code GGW20 for 20% Stay Above Nutrition products (US & Canada)!You can check them out hereSHOP OUR MERCH HERE

Experience gentle continuous rainfall designed to promote deep sleep, peaceful meditation, and ultimate relaxation with soothing nature sounds that calm the mind and body. Perfect for creating a tranquil ambiance to help you sleep better and unwind naturally.

Listen to the full show podcast of The Continuous Call Team.See omnystudio.com/listener for privacy information.

Luang Por Sumedho gave this Dhamma talk on 7 July 2026 at Amaravati Buddhist Monastery in the UK. The post Continuous Reflection on “It's Like This” Brings Us to Pure Consciousness appeared first on Amaravati Buddhist Monastery.

Luang Por Sumedho gave this Dhamma talk on 7 July 2026 at Amaravati Buddhist Monastery in the UK. The post Continuous Reflection on “It's Like This” Brings Us to Pure Consciousness appeared first on Amaravati Buddhist Monastery.

Listen to the full show podcast of The Continuous Call Team.See omnystudio.com/listener for privacy information.

In this episode of The Friday Habit, Mark welcomes back Bryan Clayton, founder and CEO of GreenPal, the on-demand lawn care marketplace connecting homeowners with local lawn care professionals across the country.After successfully building and selling a $10 million landscaping company, Bryan started over from scratch with one goal: build a company that could eventually thrive without him. Over the past decade, GreenPal has grown into a nationwide platform serving hundreds of thousands of users every week—all without outside funding.Bryan shares the lessons he's learned from scaling a technology company, building systems that outlive the founder, leveraging AI to move faster, and creating a business that is designed to be sold from day one. He also opens up about founder identity, delegation, personal health, and why working on yourself is just as important as working on your business.If you're building a service business, SaaS company, or simply trying to create something that doesn't depend entirely on you, this conversation is full of practical insights.Key TakeawaysBuild your business to function without you.Systems create freedom; hustle alone does not.Delegate only after you understand the work yourself.AI amplifies experience—it doesn't replace it.Track inputs, not just outcomes.Every dollar spent should have a measurable return.Your business should become stronger while you're away.Hire naturally motivated people instead of trying to motivate the wrong ones.Protect your health while building your company.Continuous learning is one of the greatest competitive advantages.Notable TopicsBuilding GreenPal After Selling a $10 Million CompanyBryan explains why he started over after selling his landscaping business and how he intentionally designed GreenPal to be a scalable, sellable company from day one.The Difference Between a Lifestyle Business and a Sellable BusinessMany entrepreneurs accidentally build themselves a job. Bryan discusses the mindset shift required to create a company that has value beyond its founder.Running Your Business From a SpreadsheetEvery expense should produce a return. Bryan shares why emotional decision-making can reduce business value and how disciplined financial thinking creates long-term growth.AI Is a Force Multiplier—Not a ReplacementAI dramatically increases productivity, but only when paired with real-world experience and deep understanding of your craft.Delegation Done RightOne of the biggest mistakes founders make is delegating too early. Bryan explains why you should first understand a process before handing it off.Working In, On, and On YourselfAs businesses mature, founders should gradually shift their time from daily operations toward strategy, leadership, and personal growth.Building a Team That Doesn't Need MotivationRather than trying to motivate employees, Bryan focuses on hiring naturally curious and self-driven people who already love the work they do.Health and Business PerformanceAfter reaching nearly 300 pounds while building GreenPal, Bryan realized his physical health directly impacted his leadership and decision-making.ChaptersWelcome Back Bryan ClaytonThe Growth of GreenPalScaling From Startup to Nationwide PlatformThe Three Growth Buckets: Get, Keep, GrowLeading a 300,000-User MarketplaceHow AI Is Changing EntrepreneurshipWhy Experience Still MattersBuilding a Business to SellLifestyle Business vs. Sellable BusinessRunning Every Decision Through ROIFounder Identity After Selling a CompanyBuilding Systems That Outlast YouWorking In the Business vs. On the BusinessDelegating at the Right TimeAvoiding Founder FOMOHiring Motivated PeopleLessons From Personal HealthLearning Through YouTube and AIMonday Morning ChallengeFinal ThoughtsResources & LinksLearn more about GreenPalConnect with Bryan ClaytonVisit TheFridayHabit.comDownload the free Friday Habit System guideMemorable Quotes"You don't really have a business until you can leave for a month and come back stronger.""Run your business from a spreadsheet—not from your emotions.""AI is a force multiplier for people who already know what they're doing.""Track the inputs, not the outputs.""Hire motivated people. Your job is simply not to demotivate them.""Work in the business, then on the business, and finally on yourself."Monday Morning Action ItemSpend the next 30 days intentionally learning one skill that will move your business forward.Use AI tools, online courses, podcasts, and YouTube to immerse yourself in that topic every day. Focus on mastering one area that removes a bottleneck in your business, then immediately apply what you've learned.

In this episode, Dr. Karen Litzy and Dr. Tom Walters explore the current state of the physical therapy profession, weighing the challenges with the bright opportunities ahead. If you're a PT student, new graduate, or seasoned professional, this discussion offers valuable insights on navigating growth, risk-taking, and innovation.  Key Topics Covered:  The contrasting narratives about PT career prospects—debunking myths and highlighting opportunities Factors influencing PT career decisions, including debt, job satisfaction, and entrepreneurial ventures Strategies for students and new grads to create diverse career pathways beyond traditional settings How social media has transformed PT education, branding, and income streams over the past decade The importance of trust, clarity, and confidence in communication for professional success and patient care The influence of industry challenges such as insurance constraints and healthcare system limitations Practical advice for building a cash-based practice while managing burnout The potential for social media and content creation to elevate the entire PT field The role of patient education, telehealth, and hybrid models in future PT delivery  Timestamps:  00:00 - Introduction: Is the PT profession broken or full of opportunity? 02:21 - Dr. Walters' take on whether PT is a broken system or ripe for growth 04:35 - How social media has changed career possibilities for PTs 05:42 - Advice to students and new grads about seeking opportunities and taking risks 06:47 - Personal stories of career leaps and the importance of thinking beyond traditional roles 08:48 - Building trust and clarity through communication for confidence and success 10:45 - Overcoming fears related to further education and career shifts 11:29 - Reflecting on the time investment of education and professional growth 13:00 - How teaching and social media experience have contributed to Dr. Walters' success 15:23 - The importance of consistency and long-term effort in building a following 16:52 - The evolution of social media content from early days to now 18:38 - Opportunities outside insurance-based clinics and how to leverage clinical experience 20:28 - Guiding students in career planning and the value of entrepreneurial thinking 22:28 - Strategies for transitioning from employment to ownership through small steps 24:13 - Can PT influencers replicate a successful model? Is it achievable for most? 26:26 - Social media's impact on professional visibility and community growth 27:55 - How content creation can elevate the PT profession as a whole 29:35 - The balance between hands-on care and digital/remote PT services 31:09 - The importance of rapport and human connection in healing 32:18 - Possibility of hybrid models combining in-person and remote PT 34:01 - The potential of technology to supplement, but not replace, hands-on care 36:35 - Addressing platform technical issues during the episode 38:36 - Final thoughts on building the future of PT and embracing new opportunities 39:31 - Advice for prospective DPT students on exploring career paths early 41:35 - The importance of financial literacy and budgeting education in PT programs 43:28 - Personal stories about student debt and managing educational costs 44:11 - Trust, clarity, and confidence as pillars for PT success and patient relationships 47:09 - How authenticity and communication foster trust within and outside the PT profession 50:22 - Review of what should be emphasized in PT curricula to prepare future practitioners 52:39 - The impact of viral content and strategic content creation on career trajectory 54:59 - Is the healthcare system broken or an opportunity? Dr. Walters' nuanced view 56:41 - Navigating personality types and finding the right work environment 56:55 - Practical tips for maintaining health, wealth, and intelligence in a PT career Final Takeaways:  The PT industry is both challenged and full of opportunity—proactive, strategic, and authentic efforts can help shape its future. Embracing digital tools, building trust, and diversifying skills are keys to long-term success. Continuous learning, honest communication, and openness to entrepreneurship are essential for growth in today's evolving landscape.  Thank you to Dr. Tom Walters for sharing his experience and insights. For more resources, courses, and his book, visit Rehab Science. Remember, your career trajectory depends on intentional choices, resilience, and innovative thinking—so take that leap! More About Dr. Walters Dr. Tom Walters is a board-certified orthopedic physical therapist who specializes in the treatment of pain and movement disorders. He is the founder of Rehab Science and has spent almost two decades teaching people about human movement, pain science, and how to effectively recover from injury. In addition to running his own clinical practice, Tom served as a full-time undergraduate kinesiology professor for eight years, teaching human biomechanics, therapeutic exercise, and pain science. He received his Doctor of Physical Therapy degree from Chapman University and completed a residency in orthopedic manual physical therapy and a fellowship in lower quarter functional biomechanics. Through his books, social media platforms, and the Rehab Science membership, Tom has helped millions of people better understand their bodies and take control of their recovery. Find more of Tom's content on Instagram and YouTube @rehabscience Resources from this Episode: Tom's Website: Rehab Science Tom on Instagram Tom on YouTube The Rehab Science Book Series The Rehab Science Membership   Jane Sponsorship Information: Book a one-on-one demo here Front Desk @ Jane Mention the code LITZY1MO for a free month   Follow Dr. Karen Litzy on Social Media: Karen's Twitter Karen's Instagram Karen's LinkedIn Subscribe to Healthy, Wealthy & Smart: YouTube Website Apple Podcast Spotify Stitcher iHeart Radio

The greatest competitive advantage any organization has is its people.  Amanda Van Der Heiden, Chief Learning Consultant and Coach at Global Talent Development Solutions (GTDS) shares practical strategies for aligning people with business strategy, navigating change with confidence, and cultivating leaders who inspire trust, curiosity, and continuous growth. From designing leadership development with intention to understanding why employees resist change and how emotional intelligence fuels lasting success, Amanda offers actionable insights that every leader can apply immediately.    Highlights include:   Intentional leadership design with measurable business results.   Continuous growth and why today's leaders must continuously upskill to remain effective in an AI-driven and rapidly evolving workplace.   Amanda shares the inspiration behind her free resource, 'Goal Development Participant Workbook,' to help turn aspirations into lasting results. Visit goGTDS.com    Additional resources include, The Coaching Habit by Michael Bungay Stanier and Amanda's book, Set Yourself Up to Excel: A Practical Guide to Goals, Habits and Success   Let's Talk Tuesday with GTDS podcast.   When leaders intentionally invest in learning, communicate with purpose, and develop others, they design culture where both people and businesses thrive.  Download now and continue to strengthen your leadership, empower your teams, and create meaningful organizational impact.      Timestamps: Why Resist Change 12:26 Skill Up 18:46 Goal Development Resource 24:39

In Episode 110 of the Cybersecurity Readiness Podcast Series, Dr. Dave Chatterjee is joined by Dr. Varin Khera, Co-Founder and Chief Technology Officer of SecStrike and Head of Asia Pacific at Yarix, to examine how frontier AI models have shifted the offense-defense balance in cybersecurity, and why the annual or semi-annual penetration test — long treated as a reliable baseline control — can no longer keep pace with adversaries who reason adaptively, chain misconfigurations across dozens of systems, and build their own attack playbooks in real time.Dr. Khera, who sits on both sides of the AI arms race — building EchoStrike, SecStrike's AI-driven, model-agnostic “symbiotic penetration testing” platform, while also advising enterprise clients through Yarix on how to defend against that same class of technology — walks through how frontier AI differs from the automation that preceded it. Using a locksmith analogy, he explains that older AI tools executed a fixed playbook, while frontier models construct the attack path themselves, discovering and exploiting misconfigurations a once-a-year human-led test would never have the time or reach to find. The conversation details the architecture behind EchoStrike: Crimson Nexus, a persistent, fingerprint-based knowledge engine that learns from past human decisions; the Red Engine, which orchestrates and executes validation actions; Recon, which continuously maps external attack surfaces; and the patent-pending Adaptive Threat Validation (ATV) engine that ties the components together and escalates high-judgment decisions to a human reviewer before any high-impact action is taken.Analyzed through Dr. Chatterjee's Commitment–Preparedness–Discipline (CPD) Framework, the episode also addresses how security leaders should frame the case for continuous validation to the board — not as a technology purchase, but as a decision about whether to close a known and growing risk — and closes with a rapid-fire exchange on the misconceptions, governance gaps, and accountability questions defining this next phase of AI-driven offensive and defensive security.To access and download the entire podcast summary with discussion highlights - https://www.dchatte.com/episode-110-when-the-attacker-builds-the-key-frontier-ai-and-the-future-of-continuous-penetration-testing/Connect with Host Dr. Dave ChatterjeeLinkedIn: https://www.linkedin.com/in/dchatte/ Website: https://dchatte.com/Books PublishedThe DeepFake ConspiracyCybersecurity Readiness: A Holistic and High-Performance ApproachArticles & Cases PublishedChatterjee, D. (2026). The Cryptographic Reckoning: Why Quantum Readiness Begins with Agility, Not Algorithms, The INFORMS Analytics Magazine, June 26, 2026Chatterjee, D. (2026). The New Digital Fragility: How AI-Enhanced Cyber Threats Are Reshaping Operational Resilience, The INFORMS Analytics Magazine, March 4, 2026Chatterjee, D. (2026). Root: Automating the Remediation Gap, Ivey Publishing, Jan 7, 2026.Ramasastry, C. and Chatterjee, D. (2025). Trusona: Recruiting For The Hacker Mindset, Ivey Publishing, Oct 3, 2025.Chatterjee, D. and Leslie, A. (2024). “Ignorance is not bliss: A human-centered whole-of-enterprise approach to cybersecurity preparedness,” Business Horizons, Accepted on Oct 29, 2024.Isik, O., Chatterjee, D., and Lourenco, D.A. (2024). “Getting Cybersecurity Right,” California Management Review — Insights, Accepted for Publication, July 8, 2024. Chatterjee, D. (2023). “Mission critical – How American Cancer Society successfully and securely migrated to the cloud amid the pandemic,” I by IMD, March 13, 2023.Chatterjee, D. (2022). “Preventing security breaches must start at the top,” I by IMD, September 28, 2022, Institute for Management Development, Lausanne, SwitzerlandChatterjee, D. (2022). “Making Cybersecurity Readiness Mainstream,” Executive Blog Post, NETSPI, March 1, 2022Benz, M. and Chatterjee, D. (2020). “Calculated Risk? A Cybersecurity Evaluation Tool for SMEs,” Business Horizons, available online from May 4, 2020Chatterjee, D. (2019). “Should Executives Go To Jail Over Cyber Attacks,” Journal of Organizational Computing and Electronic Commerce, Vol 29, Issue 1, pp. 1-3.Abraham, C., Chatterjee, D., and Sims, R. (2019). “Muddling through cybersecurity: Insights from the U.S. healthcare industry,” Business Horizons, July 2019.

Clinical trials provide critical answers, but they're only part of the story. In this episode of The Health Pulse, Aaron Berger and Dr. Massoud Toussi, real-world evidence executives at United BioSource (UBC), explore how the life sciences and health care ecosystem is moving beyond isolated snapshots toward continuous evidence generation. They discuss how organizations are building a more complete picture of treatment effectiveness, safety and patient outcomes in the real world. Electronic health records, connected devices, patient-centered research and global evidence networks are making that possible.Making this vision a reality requires data quality, interoperability, regulatory alignment and trusted evidence generation. AI is playing a growing role in research and analysis. It is creating new opportunities to accelerate evidence generation and uncover insights at a scale previously impossible. Ultimately, they share an optimistic vision for the future: one where real-world evidence, AI and modern data ecosystems help us learn faster, answer more questions and accelerate scientific discovery for patients who need new therapies most. 

Immerse yourself in gentle continuous rainfall with soothing ambient sounds designed to enhance sleep, meditation, and deep relaxation. Let the calming rhythm of natural rain wash away stress and create a peaceful atmosphere for rest and mindfulness.

Thomas explains how German handles continuous actions, something that doesn't come with its own dedicated tense the way English '-ing' forms do. He also introduces the increasingly popular am-progressive: 'Wir sind gerade am Entspannen', 'Gestern waren wir am Wandern'. It's a handy shortcut that also means one less set of conjugation rules to worry about.➡️ Click here to watch the video version of this episode.➡️ Get free mini-lessons and language tips every week by signing up to our newsletter: https://coffeebreaklanguages.kit.com/newsletter Hosted on Acast. See acast.com/privacy for more information.

“How long have you been learning English?” “¿Cuánto tiempo has estado estudiando inglés?” Quizás llevas meses, quizás años. Quizás estás dando tus primeros pasos, o quizás ya trabajas en un país de habla inglesa y cada día es una nueva lección. Sea cual sea tu situación, hoy vas a aprender un tiempo verbal que te va a servir para hablar de todo eso: el presente perfecto continuo. Es el tiempo que usamos cuando queremos contar lo que hemos estado haciendo — los logros, el progreso, las nuevas amistades, los retos superados. Y lo vamos a practicar a través de algo muy concreto: la experiencia de trabajar en dos de los empleos más comunes entre inmigrantes hispanohablantes — caregiver y cook. ¡Empecemos! Recuerda que todos los recursos para este episodio, incluyendo la transcripción, la tabla de vocabulario y ejercicios para repasar el aprendizaje, están disponibles en nuestro sitio web.  Haz clic en este enlace para ver todos los recursos para este episodio:   https://inglesdesdecero.ca/272 ----- Dale “me gusta” a nuestra página en Facebook: https://www.facebook.com/inglesdesde0/ ----- Síguenos en Instagram: https://www.instagram.com/ingles.desde.cero/ ----- Suscríbete en YouTube: https://www.youtube.com/@inglesdesdecero145 ----- Encuéntranos en Pinterest: https://es.pinterest.com/inglesdesdeceroca/ ----- Aprende inglés con nativos que se formaron en su enseñanza. ¡Visita nuestro sitio web, https://inglesdesdecero.ca/ para inscribirte y seguir todas nuestras lecciones! No dejes pasar esta oportunidad con Shopify y regístrate para un período de prueba por solo un dólar al mes en shopify.mx/desdecero Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

To watch a video version of this podcast, click here: https://youtu.be/vwDEHoyPgG4 In this episode, Reuben Saltzman, Tessa Murry, and Eric Houseman share some of the most memorable and unusual service inquiries they've encountered in the home inspection industry. From clients complaining about issues already documented in the inspection report to hidden defects concealed behind artwork and inaccessible attics, the conversation explores the challenges inspectors face when managing expectations after a home purchase.TakeawaysManaging client expectations before, during, and after an inspection is critical to customer satisfaction.Many service inquiries originate from misunderstandings about what a home inspection can and cannot uncover. Home inspectors are not required to move personal property, remove artwork, or access concealed areas that are not readily accessible.Hidden defects behind furnishings, artwork, insulation, or inaccessible spaces often become points of contention after a sale. Contractor opinions provided after an inspection do not automatically mean the inspector made a mistake. Home inspectors are not code inspectors, and older homes often contain conditions that would not meet current codes but are still functioning. Strong documentation, photographs, videos, and clear report writing help protect inspectors when questions arise later. AI can be a valuable quality-control tool by reviewing inspection reports for inconsistencies, missing comments, and reporting standards.Structure Tech has begun implementing an AI-powered report review process using uploaded procedures, style guides, and inspection requirements. Even experienced inspectors occasionally miss obvious issues, which can impact client confidence in the report. Showing up, listening, and communicating directly with clients often resolves concerns better than lengthy back-and-forth correspondenceMost home inspection complaints are relatively rare, especially when a company has strong training, processes, and quality control systems in place. A dedicated service manager can help objectively handle complaints without the emotional attachment inspectors may have to their own reports. Building trust through transparency and responsiveness is often more important than determining who is right or wrong.Continuous improvement through technology, training, and customer feedback strengthens both inspection quality and customer experience. Chapters00:00 Introduction and Weather Updates02:43 Inspection Fuel Conference and Sponsor Spotlight04:19 Listener Feedback on Cosmetic Defects and Reporting Standards09:54 Real Estate Agents, Buyer Expectations, and New Construction Inspections11:31 Introducing AI-Powered Home Inspection Report Reviews18:41 Free vs. Paid AI Tools and Early Results19:54 Crazy Complaint #1: The Client Who Never Opened the Report22:23 Crazy Complaint #2: Garage Door Opener That Was Actually Locked26:24 Crazy Complaint #3: Hidden Drywall Damage Behind Artwork29:17 When Does a Refund Make Sense?32:57 Complaint Statistics and Customer Service Insights35:12 Crazy Complaint #4: Hidden Attic Access and Undisclosed Structural Repairs42:46 Crazy Complaint #5: Water Heater Code Upgrade Disputes46:02 Crazy Complaint #6: "Illegal" Plumbing Connections and Contractor Claims50:20 Standing Behind Accurate Findings and Industry Expertise53:06 Crazy Complaint #7: Cracked Heat Exchanger Discovered After Closing55:03 Lessons Learned and Final Thoughts on Service Inquiries56:17 Closing Remarks and Contact Information

Every shop owner knows the workforce problem by heart. The trades got denigrated, tech ed programs got gutted, and a whole generation of kids never found out that a manufacturing career was even an option. We have all heard that speech a hundred times. What we rarely hear is a practical, repeatable way to actually fix it. In this GenCNC series episode, we sit down with two people who built one. Dave Hataj is the second-generation owner of Edgerton Gear and the founder of Craftsman with Character, a semester-long program that pairs job shadowing with a character and worldview curriculum. Courtney Silver runs Ketchie in North Carolina, where she and her husband Andy took Dave's curriculum and launched their own version, Opportunity Knocks. The through line is simple. Kids are starving for two things, purpose and community, and the shop floor is one of the best places on earth to give them both. Dave shares how a 2014 experiment with ten students grew into a Navy-funded program running across multiple states. Courtney shows why it is far less of a lift than you would think, roughly two hours a week for fourteen weeks with a handful of students. We also get into the payoff owners do not expect. Reverse mentoring from young employees, a culture where machinists suddenly want to be on camera, productivity that climbs year over year, and stories like Miguel, a first-generation immigrant who fell in love with the trade, worked six to ten in the morning before school, and just bought his first home. If you have ever done a plant tour for a school and wondered whether it actually mattered, this one reframes the whole thing. As Dave puts it, we make gears as a vehicle to invest in people. What's Covered in this Episode (2:48) Dave on the origins of Craftsman with Character as a second-generation owner of Edgerton Gear (7:02) The Navy's two million dollar, three-year contract to take the program national (8:30) Inside the Craftsman Code, from I'm not the center of the universe to the world needs me (10:30) Why you need to head to Kennametal's booth at IMTS (11:37) Courtney on discovering Dave's work during COVID and launching Opportunity Knocks (15:25) Why it's not a big lift: twenty team members, six students, two hours a week (17:11) Continuous improvement: field trips, supply chain tours, and a Mastercam capstone part (23:31) Reverse mentoring and the young energy that re-energizes the team (24:40) Twelve years of rising productivity from more engaged, purpose-driven employees (26:17) The DN Solutions install video shot by interns (and a new content culture) (29:26) Breakfast, gratitude, and the lesson on showing up on time (30:30) Turn website visitors into buyers with Navu's AI chat (31:42) Learn more about Miguel, a first-generation immigrant who fell in love with the trade (34:10) Break room culture and crews that stick around after shift (37:01) From feeling less than to finding a joyful place to belong (39:16) The mindset flip: we make gears as a vehicle to invest in people (40:07) Get in the room: the Job Shops Workshop and reception at IMTS (40:58) Miguel buying his first home, and the ripple effect on a community (42:56) How to start, facilitator and mentor training, plus a track for shops without a school partner (45:00) Manufacturers as an army that can heal society by being the hero for one kid (48:52) Practical encouragement to start now Resources Mentioned Craftsman with Character Edgerton Gear Ketchie The Craftsman Code by Dave Hataj Machine Shop Mastery, Episode 106 with Dave Hataj Kennametal Navu IMTS Oscar Mike Foundation Connect with Dave Hataj & Courtney Silver Dave Hataj, Edgerton Gear Craftsman with Character Courtney Silver, Ketchie Connect with MakingChips Website On Facebook On LinkedIn On Instagram On Twitter On YouTube

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

Listen to the full show podcast of The Continuous Call Team.See omnystudio.com/listener for privacy information.

Grant shares the real origin story—growing up around dirt bikes and construction with his dad, the moment a tensioned bungee cord sunk into his father's hand, and how that gap in the market stuck with him for years. After becoming a dad himself and feeling limited in the trades, he dug the old "Grant Strap" idea out of his phone notes during COVID and decided to actually build it. They dig into the messy reality of inventing and launching a physical product: the prototype phase with Home Depot parts and Alibaba springs, finding the right tension (about 120 lbs on the first model), avoiding the "mailbox money" invention scammers, making and selling the first ugly versions yourself, and why "people don't buy ideas—they buy products." Grant explains how the spring-loaded design works, the continuous polyester limiter band for safety and load rating, why it stays tight when loads shift or compress (coolers, cardboard, UTVs, Jeeps), and the three models (1", 1.5", and the 2" Pro with ratchet) that cover everything from college moves to Moab trailers. The conversation also covers the less glamorous side of business—patent threats that turned into nothing, copycats, why building a strong brand and customer service is better protection than a patent alone, working-load math that goes beyond just vehicle weight, all-weather durability, proper care for lifetime performance, pack sizes, lengths, and extensions. Grant talks about his Invent With Me podcast (helping inventors avoid the same pitfalls he hit) and what's next for TorkStrap: more trade shows, social content, and partnerships rather than diluting the core product. If you've ever cursed a ratchet strap in the rain, watched a load loosen on the highway, or thought about turning a simple idea into a real business, this one's for you. **Timestamps** 00:00 AI Sign Language for Deaf Viewers 00:41 Early Inspiration for Better Cargo Straps 02:36 Childhood Prototypes Reveal Market Gap 04:10 COVID Motivation to Launch Business 07:04 Inventor Licensing Pitfalls and DIY Sales 09:48 Prototyping, Sampling, and First Customers 11:52 Designing the Spring-Loaded Torque Strap 14:06 How Torque Straps Work and Their Benefits 22:48 Creative Branding Ideas for Torque Straps 24:05 Facing Competition and Patent Infringement 27:33 Defending the Product Against Patent Claims 28:12 Commitment to Quality and Customer Support 29:10 Competing with Copycats and Maintaining Reputation 30:05 Encouraging Ethical Purchasing 30:58 Patent Vulnerabilities and Enforcement Challenges 35:53 Introducing the Invent With Me Podcast 37:17 Calculating Load Ratings and Strap Sizing 44:28 All-Weather Durability and Maintenance 45:30 Spring Longevity and Packaging Options 47:02 Pricing Packs and Managing Expectations 48:04 Customizing Strap Lengths for Use 49:01 Strap Longevity and Care Guidelines 50:50 Subscription Idea Mentioned 51:04 Social Media Presence and Promotion 54:35 Upcoming Marketing Focus for Torque Strap 56:09 Shift from Engineering to Marketing 56:40 Humorous Product Concepts and April Fools 57:08 Closing Remarks and Podcast Promotion 57:42 Final Thoughts and Gratitude **Key takeaways** - Just pull—spring tension adapts to load shift or compression so you rarely (or never) have to stop and re-tighten. - Continuous polyester strap carries the actual load rating; the spring is the adaptive element with a limiter. - Three models cover light-to-heavy needs, including a 2" Pro with ratchet for vehicles and heavier cargo. - Lifetime warranty when treated with respect; average "cheap strap" life is ~2 years because people leave them in the bed. - Best protection in business isn't always the patent—it's being the guy who answers the phone, honors the warranty, and shows up every day. Check out TorkStrap: Website: https://torkstrap.com Social (all platforms): @tork_strap Invent With Me podcast: Search "Invent With Me" on YouTube, Spotify, or Apple TorkStrap M500 14' x 1'' Self-Tensioning Tie Down Straps (4-Pack) https://www.amazon.com/TorkStrap-M500-Self-Tensioning-Straps/dp/B0B9J1JW7M TorkStrap HD750 14' x 1.5'' Self-Tensioning Tie Down Straps https://www.amazon.com/TorkStrap-Self-Tensioning-Straps-Patented/dp/B0CHGDFL2W TorkStrap PRO 25' x 2'' Self-Tensioning Ratchet Straps https://www.amazon.com/TorkStrap-Tensioning-Ratchet-Straps-Strength/dp/B0F1N84TY1 TorkStrap HD750 Bundle with Carry Bag https://www.amazon.com/TorkStrap-Bundle-Spring-Loaded-Straps/dp/B0CQT6ZTKX Every purchase through these links helps support the Jeep Talk Show at no extra cost to you. For more recommended Jeep gear, tools, and show favorites, visit https://jeeptalkshow.com/amazon. Visit our website: https://jeeptalkshow.com/ Watch/Listen on Spotify https://jeeptalkshow.com/spotify Join our Discord Server: https://jeeptalkshow.com/discord Subscribe to our newsletter: https://jeeptalkshow.com/newsletter Help Support the show via Patreon: https://jeeptalkshow.com/patreon

Did you start taking progesterone and now confused if you are supposed to take it every day or only for two weeks? Or maybe you don't have a uterus anymore, and you keep going back and forth on whether you should take it anyway or just leave it? You heard there are both risks and benefits and not sure what to do? We cover: How to take progesterone if you're cycling What to do when progesterone gives you two great weeks of sleep, but told to stop for two weeks. Continuous versus cyclic, and how to tell which one your body would prefer The latest information about progesterone and cancer, heart, and dementia risk What I would personally do about progesterone if I didn't have a uterus   MENOPAUSE TRACKER https://freebie.hackmyage.com/menopause-symptom-tracker MENOPAUSE SYMPTOM GUIDE https://freebie.hackmyage.com/menopause-symptom-chart Give thanks to our sponsors: Try Vitali skincare. 20% off with code ZORA here - https://vitaliskincare.com Get Primeadine spermidine by Oxford Healthspan. 15% discount with code ZORA ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠here⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ - https://www.oxfordhealthspan.com/ZORA Get Mitopure Urolithin A by Timeline. 20% discount with code ZORA at https://timeline.com/zora Try MitoQ Hormonal Metabolic Control. 15% off with code ZORA http://www.mitoq.com/hackmyage Join the Hack My Age community on: YouTube: https://youtube.com/@hackmyage YouTube Clips: https://youtube.com/@hackmyageclips Facebook Page: ⁠⁠⁠⁠⁠@⁠Hack My Age⁠     Facebook Group: ⁠⁠⁠⁠⁠⁠@⁠Biohacking Menopause⁠⁠⁠⁠⁠⁠ ⁠   Biohacking Menopause Private Women's Only Support Group (Join 7 days free): https://membership.hackmyage.com/sales-page Instagram: ⁠⁠⁠⁠⁠@⁠HackMyAge⁠    Website: ⁠⁠⁠⁠⁠⁠HackMyAge.com⁠    For partnership inquiries: https://www.category3.ca/ Some episodes of Hack My Age are supported by partners whose products or services may be discussed during the show. The host may receive compensation or earn a minor commission if you purchase through affiliate links at no extra cost to you. All opinions shared are those of the host and guests, based on personal experience and research, and do not necessarily represent the views of any sponsor. Sponsorships do not imply medical endorsement or approval by any healthcare provider featured on this podcast.

Operational leadership isn't built in the boardroom. It's built through years of solving problems, understanding people, refining processes, and making decisions that strengthen every part of an organization. The most effective leaders rarely begin at the top. They build their perspective one role at a time, gaining firsthand knowledge of how operations, customer experience, technology, supply chains, and leadership intersect. That breadth of experience often becomes their greatest competitive advantage, especially during periods of uncertainty. Today's business environment demands exactly that kind of leadership. Organizations are navigating economic shifts, changing consumer expectations, workforce challenges, emerging technologies, and increasing competition. Navigating those complexities requires more than expertise in a single discipline. It requires leaders who understand how every function of the business contributes to long-term success. Operational leadership begins with that understanding. One of the biggest misconceptions about leadership is that executives eventually outgrow operations. In reality, the strongest leaders remain closely connected to the daily realities of their organizations. They understand the challenges facing employees, the needs of customers, and the pressures experienced by business owners and operators because they've often lived those experiences themselves. That perspective creates better decisions. Rather than making assumptions from behind a desk, operational leaders recognize how changes in one area affect every other part of the business. Marketing influences operations. Operations shape customer experience. Customer experience drives loyalty. Technology impacts efficiency. Every decision creates a ripple effect throughout the organization. As Jeff Hetsel puts it, "Great leaders don't just understand one department. They understand how every part of the business works together." That philosophy has become increasingly important as organizations continue adapting to rapid change. Few industries illustrate this better than the restaurant business. The COVID-19 pandemic challenged nearly every assumption about how restaurants operated. Dining rooms closed, customer expectations changed overnight, supply chains became unpredictable, and operators were forced to rethink nearly every aspect of their businesses. While every organization faced difficult decisions, the companies that emerged strongest shared several common characteristics. They communicated frequently, adapted quickly, stayed close to their customers, and maintained strong relationships with the people responsible for executing the business every day. Communication proved especially valuable. When uncertainty increases, information becomes leadership. Organizations that communicated consistently with franchisees, employees, suppliers, and customers were often able to make better decisions because everyone understood the challenges, priorities, and direction of the business. Transparency created trust, and trust created alignment. That principle extends far beyond franchising. Whether leading a small business or a global organization, communication remains one of the most effective operational tools available. People perform better when they understand not only what is changing, but why those changes matter. Operational leadership also requires the discipline to continually evaluate how technology supports the customer experience. Artificial intelligence, automation, digital ordering, customer relationship management systems, and advanced analytics are reshaping nearly every industry. Businesses that ignore these innovations risk falling behind. At the same time, technology should never become a substitute for genuine human connection. Instead, the most successful organizations use technology to remove friction. Automating repetitive tasks allows employees to focus on serving customers, solving problems, and building relationships. Rather than replacing people, technology should create more opportunities for meaningful interactions. This balance will likely define the next generation of business leadership. Consumers increasingly expect convenience, speed, and personalization. They also continue to value authenticity, trust, and personal service. Organizations capable of delivering both will create stronger customer loyalty and long-term competitive advantages. Continuous learning is another defining characteristic of operational leadership. Business landscapes evolve too quickly for leaders to rely solely on past experience. Markets shift. Competitors innovate. Customer preferences change. The leaders who continue growing are those who remain curious enough to keep learning. Books remain one of the simplest ways to develop that perspective. While digital content provides quick answers, books offer something different: depth, context, and thoughtful analysis. Many accomplished executives continue to make reading a priority because it exposes them to new ideas, leadership philosophies, and strategies that can be applied long before competitors recognize the opportunity. That mindset reflects another simple but powerful philosophy. "You have to earn your job every day." Leadership is never permanent. Every day presents new opportunities to improve processes, strengthen teams, create value, and serve customers more effectively. The strongest leaders understand that success yesterday guarantees nothing tomorrow. They remain students of their industry, constantly asking better questions and looking for smarter ways to operate. Perhaps the most overlooked aspect of operational leadership is service. Leadership is often associated with authority, decision-making, and accountability. Those responsibilities certainly matter. Yet the organizations that consistently outperform their competitors often embrace a different philosophy. They view leadership as service. Serving employees. Serving franchisees. Serving customers. Serving communities. That perspective influences every decision throughout the organization. As Hetsel explains, "Being great is anything you do in the service of others." It's a simple statement, yet it captures an essential truth about sustainable business growth. Organizations succeed when the people inside them succeed first. Strong leaders remove obstacles instead of creating them. They build systems that support consistency. They communicate with transparency. They embrace innovation without abandoning the human experience that customers value most. Operational leadership is not about knowing every answer. It's about understanding the business well enough to ask better questions. It's about remaining curious after decades of experience. It's about recognizing that growth depends on people just as much as processes. Most importantly, it's about never becoming disconnected from the customers, employees, and partners who make long-term success possible. Business will continue to evolve. Technology will continue advancing. Customer expectations will continue changing. The organizations best positioned for the future will be led by individuals who understand operations from the ground up, lead through service, embrace continuous learning, and never lose sight of the people behind every business decision. About Jeff Hetsel Jeff Hetsel is President of Cicis Pizza and JMC Restaurant Distribution, bringing nearly 40 years of restaurant and franchise leadership experience. A Certified Franchise Executive, Jeff began his career with Cicis in 1992 as a restaurant manager and has since served in leadership roles spanning operations, franchise development, real estate, construction, distribution, and executive management. His hands-on experience across virtually every aspect of the business has helped guide the brand through significant industry change while supporting franchisees, strengthening operations, and positioning Cicis for continued growth. About Ford Saeks Ford Saeks is a Business Growth Accelerator who has generated more than a billion dollars in sales worldwide by helping businesses attract loyal customers, increase visibility, and accelerate growth. As President and CEO of Prime Concepts Group, Inc., Ford has founded more than ten companies, authored eleven books, earned three U.S. patents, and advised organizations ranging from startups to Fortune 500 companies. A recognized expert in business growth, customer acquisition, leadership, franchising, marketing, and AI-driven business strategies, Ford helps business owners and leaders identify opportunities, improve performance, and achieve sustainable results. Learn more at ProfitRichResults.com and watch Fordify LIVE at Fordify.tv

In this episode of the Prolonged Field Care Podcast, Dennis sits down with J.R. Pickett — unpack the controversial and high-stakes topic of what used to be called excited delirium.They dig into the history of the syndrome (Bell's mania, acute exhaustive mania, agitated delirium), why major organizations including ACEP, ACMT, and the National Association of Medical Examiners have rejected the term, and the preferred modern language: hyperactive delirium with severe agitation. The conversation covers real-world presentation, the physiologic cascade that can lead to sudden cardiovascular collapse, the critical differences between a contained hospital environment and the uncontrolled street or austere setting, and the hard lessons from the Elijah McClain case.J.R. walks through practical decision-making for EMS and tactical medics: when de-escalation is possible, when sedation becomes necessary, why intramuscular ketamine remains the most forgiving and rapid option for the violently agitated patient, how to prepare for the predictable risks (brief apnea, loss of airway protection, metabolic derangement), and why continuous medical eyes-on monitoring after sedation is non-negotiable. They also address the dangerous intersection of law enforcement and medical care, the myth of “if they can talk they can breathe,” and the growing criminalization of medical decision-making that threatens providers' willingness to engage.Key TakeawaysThe condition is a true medical emergency with historically high mortality, even without restraint or intervention.Engagement ability is a practical field litmus test: if the patient cannot be redirected or answer basic questions, rapid intervention is usually required.Ketamine's wide therapeutic index and rapid IM onset make it the preferred agent for violent agitation when IV access is impossible — but it is not risk-free.Sedation is a procedure. Have airway equipment, monitors, and a clear team plan ready before the drug is given. Continuous medical provider eyes-on is mandatory in the early phase.“If you can talk, you can breathe” is dangerous teaching. Treat complaints of inability to breathe seriously.Noble intent + thorough preparation is the best defense against both bad outcomes and the growing criminalization of medical care.Chapters02:45 – What is (or was) excited delirium? History, physiology, and why the term is being abandoned09:30 – Real-world presentation vs. “just being a jerk” and the challenge of the uncontrolled environment15:20 – Elijah McClain case and the broader controversy around restraint, force, and medical justification21:00 – Causes of severe agitation and the difficulty of sorting them in the field26:45 – Clinical clues and the “can I engage?” litmus test32:10 – The physiology of sudden collapse: acidosis, rhabdomyolysis, and the danger of sudden quiet37:40 – “I can't breathe” and why that teaching is hazardous45:50 – Ketamine deep dive: dosing, therapeutic index, risks, and why it is still the safest rapid option55:20 – Comparison with benzodiazepines and antipsychotics; timing matters01:01:00 – Treating sedation like a procedure: airway readiness, monitoring, team roles, and continuous eyes-on01:10:30 – Police vs. medical roles, the myth of walking away, and the duty to act01:18:00 – Criminalization of medical care and final thoughts on honorable intentFor more content, go to ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠www.prolongedfieldcare.org⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Consider supporting us: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠patreon.com/ProlongedFieldCareCollective⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ or ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠www.lobocoffeeco.com/product-page/prolonged-field-care⁠⁠⁠⁠

If you or someone you know is on a GLP-1 medication like Ozempic, Wegovy, or Mounjaro and the results have plateaued, or the weight is starting to creep back, this episode has the explanation you have been looking for and it is probably not what your doctor told you. This week Susie breaks down exactly why these medications stop working for so many people, what is actually going wrong metabolically, and what to do about it instead of just increasing the dose. Then Leanne takes over for a genuinely useful segment on the Coles and Woolworths home brand products both dietitians buy every single week, which ones are just as good as the branded version, and which ones they would never swap out. In this episode: Why GLP-1 medications like Mounjaro, Ozempic, and Wegovy stop working: the two metabolic patterns Susie sees most commonly in her clients, and why increasing the dose is often exactly the wrong response Why dramatically reducing calorie intake without addressing muscle mass and exercise will always lead to a plateau, and the counterintuitive reason why eating a little more and reducing the dose can actually restart results Why muscular, active women are particularly vulnerable to GLP-1 plateaus, and how being in carbohydrate no man's land sabotages fat burning even when the medication is working properly Why these medications should never be given without dietitian support, and what to actually focus on if you want to come off them eventually without regaining the weight The Coles and Woolworths home brand products Leanne and Susie buy every single week without hesitation: rolled oats, canned legumes, passata, frozen vegetables, cottage cheese, Greek yogurt, chicken breast bites, marinated chicken, and more. Specific product names, aisle locations, and honest reasons why Why the Coles Kitchen frozen grilled vegetables and the Woolworths chicken breast bites are non-negotiables in both shopping trolleys, and the specific products in the Coles Perform and Coles Kitchen ranges that keep making it back onto the list The Aldi Gourmet Protein Bar reviewed: 10 grams of protein, 37% nuts and seeds, less than a dollar per bar, and a price-per-protein calculation that puts most supermarket bars to shame. What the ingredient list actually contains and who this bar is genuinely suited to Continuous glucose monitors: who actually needs one, why Leanne and Susie have both turned down paid partnerships with CGM companies, and the specific situations where they might genuinely be useful versus where they are more likely to create unnecessary food fear See omnystudio.com/listener for privacy information.

Continuous improvement is grounded in learning. In fact, continuous improvement necessitates a mindset that focuses on continued learning and growth and that acknowledges that there is always something new to discover. Through this learning, growth occurs, and growth facilitates improvement. As Brian Alaback, Director of Professional Learning for Escambia County Public Schools in Florida so aptly shares, “if I don't have the mindset of continuing to grow and learn, then it doesn't matter what I'm trying to improve.”   Listen as Dr. Janet Pilcher and Mr. Alaback consider how leaders model and lead the way in growing and learning to drive improvement, explore why investing in your own learning and the learning of others is crucial to continuous improvement, and discuss tactics for supporting improvement through learning. Recommended Resources:  Continuous Improvement Through Leadership, Build a Culture of Improvement  Follow Host Dr. Janet Pilcher on LinkedIn: https://www.linkedin.com/in/janetpilcher/ 

Episode: 1607 Continuous-aim firing: a diagnosis of an ill-received idea.  Today, we aim a gun from a rocking platform.

God holds the future!Do you wonder what the future holds? The Bible's record of proven correctness, extending over many centuries, validates its claim to predict the future with authority and accuracy.Support the show

Long-Form Standing Journey — Complete Practice (73 min, preview 23 min)I love sequences that just keep going — a long journey through many postures on one side before you switch and do the other. This is one of those. Rather than resetting frequently, this class asks you to remain present through a long, unfolding progression of standing postures, balance work, and transitions that build naturally on one another.The challenge isn't simply physical—it comes from remembering where you've been, staying engaged as the sequence evolves, and trusting the process without rushing toward the other side. The result is a practice that feels immersive, satisfying, and deeply rewarding, with each posture setting up the next in one continuous conversation.This is a class that rewards patience, attention, and the willingness to stay with something a little longer than usual.New class every Tuesday. 350+ full classes in the Unlimited Archive at JustGreatYoga.com

Gregory Copley details King Charles III's visit to the Isle of Man, the world's oldest continuous parliament. The King also held a private, discreet meeting with Prince Harry. Meanwhile, the UK faces rising anxiety over security for political leaders following the assassination of Ann Widdecombe. (12)1901