POPULARITY
Categories
394 episodes with Dynamo
227 episodes with Dynamo
213 episodes with Dynamo
219 episodes with Dynamo
154 episodes with Dynamo
118 episodes with Dynamo
38 episodes with Dynamo
67 episodes with Dynamo
70 episodes with Dynamo
43 episodes with Dynamo
59 episodes with Dynamo
19 episodes with Dynamo
31 episodes with Dynamo
26 episodes with Dynamo
48 episodes with Dynamo
26 episodes with Dynamo
50 episodes with Dynamo
19 episodes with Dynamo
30 episodes with Dynamo
31 episodes with Dynamo
12 episodes with Dynamo
37 episodes with Dynamo
12 episodes with Dynamo
23 episodes with Dynamo
12 episodes with Dynamo
22 episodes with Dynamo
5 episodes with Dynamo
12 episodes with Dynamo
8 episodes with Dynamo
8 episodes with Dynamo
18 episodes with Dynamo
9 episodes with Dynamo
14 episodes with Dynamo
20 episodes with Dynamo
7 episodes with Dynamo
8 episodes with Dynamo
10 episodes with Dynamo
7 episodes with Dynamo
13 episodes with Dynamo
4 episodes with Dynamo
19 episodes with Dynamo
9 episodes with Dynamo
5 episodes with Dynamo
5 episodes with Dynamo
10 episodes with Dynamo
5 episodes with Dynamo
12 episodes with Dynamo
8 episodes with Dynamo
14 episodes with Dynamo
10 episodes with Dynamo
8 episodes with Dynamo
5 episodes with Dynamo
7 episodes with Dynamo
5 episodes with Dynamo
8 episodes with Dynamo
6 episodes with Dynamo
6 episodes with Dynamo
5 episodes with Dynamo
5 episodes with Dynamo
4 episodes with Dynamo
6 episodes with Dynamo
4 episodes with Dynamo
6 episodes with Dynamo
4 episodes with Dynamo
5 episodes with Dynamo
3 episodes with Dynamo
5 episodes with Dynamo
9 episodes with Dynamo
5 episodes with Dynamo
2 episodes with Dynamo
5 episodes with Dynamo
3 episodes with Dynamo
6 episodes with Dynamo
6 episodes with Dynamo
3 episodes with Dynamo
3 episodes with Dynamo
2 episodes with Dynamo
9 episodes with Dynamo
4 episodes with Dynamo
3 episodes with Dynamo
3 episodes with Dynamo
3 episodes with Dynamo
3 episodes with Dynamo
3 episodes with Dynamo
2 episodes with Dynamo
2 episodes with Dynamo
63 episodes with Dynamo
3 episodes with Dynamo
4 episodes with Dynamo
2 episodes with Dynamo
6 episodes with Dynamo
6 episodes with Dynamo
6 episodes with Dynamo
5 episodes with Dynamo
3 episodes with Dynamo
3 episodes with Dynamo
3 episodes with Dynamo
2 episodes with Dynamo
3 episodes with Dynamo
6 episodes with Dynamo
6 episodes with Dynamo
3 episodes with Dynamo
2 episodes with Dynamo
6 episodes with Dynamo
3 episodes with Dynamo
54 episodes with Dynamo
5 episodes with Dynamo
2 episodes with Dynamo
Hour 2 of Soccer Matters with Glenn Davis! Glenn reflects on the passing of country music icon Dolly Parton ThirdDegree.net's Buzz Carrick joined the show Jon Nelson from 'Soccer Down Here' on a record breaking night for soccer in Georgia Glenn breaks down the current Dynamo situation
In this episode, Madelyn O'Farrell interviews Eesha Pathak, Senior Director of Product Management at Crusoe, about her unconventional career from software engineering and branding to leading enterprise AI at Google and now building AI-first infrastructure. They discuss Crusoe's vertically integrated, energy-first approach to cloud and data centers, capturing abundant energy (like stranded gas and renewables) and turning “electrons into intelligence” via GPUs and a custom cloud stack. Eesha explains how Crusoe's gigawatt-scale campuses and modular Spark edge deployments complement each other to deliver both scale and low-latency inference, dives into the Managed AI platform with its model marketplace, self-serve and tailored deployments, fine-tuning, and upcoming reinforcement learning, and highlights why judgment and holistic thinking are critical in product and engineering. She also unpacks Crusoe's close partnership with NVIDIA, its role in enabling physical AI and robotics with low-latency compute, a flexible data and partnership strategy, and her long-term vision of Crusoe becoming the most helpful company for people building AI, so customers can focus on their products instead of wrestling with infrastructure. Highlights from their conversation include: Eesha's Unconventional Career Path from Bosch to Google to Crusoe (0:29) Why Energy Is the Real Bottleneck for AI Infrastructure (3:30) What Energy-First Means and Bringing Compute to Abundant Power (5:04) Spark Modular Data Centers and Edge Zones Strategy (6:22) How Gigawatt Campuses and Edge Zones Work Together (8:26) Importance of Judgment in Product and Engineering Decisions (12:49) NVIDIA Partnership and Day Zero Nemotron Model Launches (15:25) Physical AI, Robotics, and Low-Latency Edge Inference (17:21) Crusoe's Data Strategy and Partnership Approach (21:52) Vision for Crusoe as Most Helpful Company for AI Builders (22:47) Closing Thoughts and Episode Wrap-Up (24:09) Dynamo Ventures is a venture firm backing founders upgrading the physical economy. As intelligence moves into critical infrastructure and technology collides with physics, industry is entering a new era of transformation - the industrial renaissance. Born from the dirt and grit of supply chains and shaped by operations, not spreadsheets, Dynamo focuses on the complex realities of building in the real world. We invest in companies transforming infrastructure, manufacturing, logistics, transportation, and the systems that power global commerce. Dynamo works closely with founders who combine ambition with a bias to action, bringing a builder mindset to venture capital through deep operational insight, systematic pressure-testing and hands-on partnership. Our purpose is simple: to back the relentless shaping the industrial renaissance. Learn more at www.dynamo.vc. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
In this episode, Turner Construction's Drew Kerr, Vice President and General Manager of xPL Offsite, explains how xPL is redefining construction through offsite manufacturing and deep supply chain integration, starting with data centers and expanding into sectors like semiconductors, healthcare, and industrial facilities. He contrasts traditional, fragmented construction with xPL's unified model that centralizes manufacturing, supply chain, and field installation to handle unprecedented demand and labor shortages. Using a rapid 30MW data center deployment as a case study, Drew shows how close coordination, prefabrication, and early procurement decisions enabled timelines that would be impossible under conventional methods. He also discusses the limits of true “manufacturing” in construction today, xPL's use of automation and robotics, and why he believes modularization will become the dominant way we build in the next century. Highlights from their conversation include: Drew's Background at Turner and Personal Story (0:54) What xPL Offsite Does and Why It Was Created (2:02) How Exponential Party Logistics Reimagines Construction Supply Chains (4:15) Industrial Renaissance, Data Centers, and Physical Economy Bottlenecks (7:25) Why Offsite Construction Is Not Yet True Manufacturing (8:23) Case Study: 30MW “Emergency” Data Center Deployment (10:36) How Integrated Teams Beat Traditional Construction Timelines (11:06) Labor Shortage, Skilled Trades, and Bringing Work to Workers (13:25) Role of Automation, Robotics, and Smart Factory Concepts (15:50) Expanding Beyond Data Centers Into Healthcare, Sports, and Industrial (19:38) Future of Construction and Shift to Modularized Building Models (21:51) Global Modularization Vision for Turner and ACS (24:10) Closing Thoughts and Episode Wrap-Up (25:29) Dynamo Ventures is a venture firm backing founders upgrading the physical economy. As intelligence moves into critical infrastructure and technology collides with physics, industry is entering a new era of transformation - the industrial renaissance. Born from the dirt and grit of supply chains and shaped by operations, not spreadsheets, Dynamo focuses on the complex realities of building in the real world. We invest in companies transforming infrastructure, manufacturing, logistics, transportation, and the systems that power global commerce. Dynamo works closely with founders who combine ambition with a bias to action, bringing a builder mindset to venture capital through deep operational insight, systematic pressure-testing and hands-on partnership. Our purpose is simple: to back the relentless shaping the industrial renaissance. Learn more at www.dynamo.vc Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
It's time to go back to the 1980s, when giant robots ruled the pop culture landscape with rocket fists, and talking about Dynamo Joe! A love letter to Japanese media like Robotech and Gundam, we talk about how the comic is mostly a successful send-up to its source material
The final hour features a special guest as Houston Dynamo legend Brian Ching calls into ITL to discuss the Dynamo, the impact of the 2026 World Cup, Houston's growing soccer culture, and more. Ching remains one of the most accomplished players in Dynamo history and is part of the 2026 Houston Sports Hall of Fame class. The show wraps with Figgy's Mixtape, featuring the latest on a local serial squatter heading back to court, comedian and content creator Damon Darling being banned from Walmart, and more wild stories from around the internet.
ITL kicks off a historic day for Houston sports radio as the guys officially welcome listeners to the NEW 95.7 The Fan, reacting to the station flip and reflecting on what the beginning of this new era means for the station, its listeners, and everyone who helped build it. The conversation turns to the Texans tight end room, as ITL discusses what Dalton Schultz and Marlin Klein could bring to the offense before Houston rap legend Paul Wall joins the show to congratulate the guys on the launch of 95.7 The Fan and celebrate the big day. ITL then goes Around The NFL, reacting to the biggest news and storylines from across the league before looking at which Texans starters could be in danger of losing their jobs as training camp position battles continue to heat up. The hour wraps with What's Popping, covering the biggest stories from sports and entertainment. With the Texans opening the preseason against the Los Angeles Chargers, the guys discuss which players they're most looking forward to watching, which roster hopefuls have the most to prove, and which position battles could begin taking shape once the lights come on. The crew then opens the floor for another edition of ITL Lunch-Time Confessions before playing Buy or Sell with the biggest sports topics and debates of the day. The historic first show on 95.7 The Fan closes with another special guest as Houston Dynamo legend Brian Ching calls in to discuss the Dynamo, the World Cup, Houston's soccer scene, and more. Finally, Figgy's Mixtape wraps up the day with the latest on a local serial squatter heading back to court, comedian and content creator Damon Darling getting banned from Walmart, and more wild stories from around the internet.
Brian Ching calls into the show to talk Dynamo soccer, the World Cup, Houston's soccer scene, and more.
Samstagabend, Flutlicht, Betzenberg – das Drumherum passt auf jeden Fall! Aber wie gehen die Teams rein ins erste Spiel der neuen Saison? Der 1. FC Kaiserslautern hat zum Auftakt einen Punkt in Wolfsburg geholt, der nicht unbedingt einkalkuliert war. Der Karlsruher SC hat mit ein bisschen Aluglück drei Punkte gegen Bielefeld geholt. Jetzt geht es für die beiden am zweiten Spieltag in eine Partie, die zwar auch nur drei Punkte bringt, aber die Weichen in Richtung erfolgreichem Saisonstart stellen kann oder eben einen ersten Dämpfer bringen. Wie stehen die Chancen für den FCK und den KSC? Wir sprechen drüber! Alle Bundesligaspiele bekommt ihr hier in voller Länge: https://www.sportschau.de/fussball/bundesliga/alle-audiostreams-der-fussball-bundesliga,audiostreams-bundesliga-uebersicht-100.html Die 2. Bundesliga gibt's hier: https://www.sportschau.de/fussball/bundesliga2/alle-audiostreams-der-2-fussball-bundesliga,audiostreams-zweite-bundesliga-uebersicht-100.html Und hier bekommt ihr den Podcast in voller Länge: https://www.sportschau.de/podcasts/sportschau-fussball-podcast Zur Doku "Alternative für Arbeiter? - Mein Kumpel wählt AfD" geht's hier: https://www.ardmediathek.de/film/alternative-fuer-arbeiter-oder-doku/Y3JpZDovL25kci5kZS80ODc4IDEyZDA0MGQxLWYwZjgtNDA0OS04NWU3LTQzNGMwMjNhYTc0NQ (00:00) Die 2. Bundesliga ist zurück(01:00) Wer holt sich den ersten Derbysieg im Südwesten? (05:02) Formcheck vor dem Frankenderby(09:04) Wolfsburg unter Hannover unter Druck?(14:03) Marcel Rapp zurück in Kiel(19:01) Dynamo mit Fanwucht gegen Darmstadt(23:40) Fortsetzung des Braunschweiger Traumstarts?(28:01) Kontert auch Osnabrück Magdeburg aus?(31:00) Berliner Weg gegen Heidenheimer Konstanz(33:49) Wer ist der Underdog auf der Alm?(37:51) Doku-Tipp: "Alternative für Arbeiter? - Mein Kumpel wählt AfD"
Soccer Matters with Glenn Davis!Glenn traveled across the US with guests from Atlanta, Seattle, and Austin to discuss how soccer in America, and MLS is changing because of the success of hosting the World CupPlus conversations on the big leagues getting started soon, the recent play of the Dynamo, and Game on for 90 on the need for transparency in sports ownership/management.
In this episode, Madelyn O'Farrell and Santosh Sankar unpack the data center boom and the idea of compute as the next utility powering an “industrial renaissance.” They explore how AI models are commoditizing, shifting value to the application layer, and draw historical parallels to industrialists like Rockefeller and Carnegie in terms of capital intensity, vertical integration, and long-lived infrastructure. The discussion dives into the biggest bottleneck (access to energy and grid capacity) along with underwhelming GPU utilization, the need for better observability and efficiency, and trends like prefab “constructuring” in data center construction. They also highlight labor and skills constraints in specialty construction, tools like Record Lens to digitize field operations, and the potential for a Foxconn-style contract manufacturer for electrical equipment. The episode closes on what excites them about founders in this space: deep problem understanding, real industrial pain points, and the ambition to build in the physical economy rather than chasing AI hype. Highlights from their conversation include: Setting up Compute as a New Utility and AI Data Center Boom (0:38) Why Compute Becomes a Utility and Implications for Trillion Dollar Tech (3:50) Drawing Parallels Between AI Infrastructure and the Industrial Revolution (6:56) Capital Intensity, Supply Chains, and Long Lived Industrial Assets (7:50) Financing Data Centers Like Power Plants and Identifying Key Bottlenecks (11:44) Energy Queue, Grid Constraints, and Alternative Generation Opportunities (12:25) Efficiency, Grid Utilization, and Rising Importance of Operational Arbitrage (15:13) Utilization, ROI vs. Dark Capacity, and Lessons from the Dot Com Era (21:23) Constructuring Trend and Prefab Manufacturing for Data Centers (26:16) Record Lens and AI Native Project Management for Grid Scale Construction (29:24) Idea of a Foxconn Model for Electrical Equipment Manufacturing (32:47) Standardization, Certification, and Cyber Risk in Grid Infrastructure (36:22) Founder Traits, Industrial Ambition, and Solving Top Three Customer Problems (38:00) Gold Rush Dynamics, Real Pain Points, and Building in the Physical Economy (41:31) FInal Thoughts and Takeaways (42:42) Dynamo Ventures is a venture firm backing founders upgrading the physical economy. As intelligence moves into critical infrastructure and technology collides with physics, industry is entering a new era of transformation - the industrial renaissance. Born from the dirt and grit of supply chains and shaped by operations, not spreadsheets, Dynamo focuses on the complex realities of building in the real world. We invest in companies transforming infrastructure, manufacturing, logistics, transportation, and the systems that power global commerce. Dynamo works closely with founders who combine ambition with a bias to action, bringing a builder mindset to venture capital through deep operational insight, systematic pressure-testing and hands-on partnership. Our purpose is simple: to back the relentless shaping the industrial renaissance. Learn more at www.dynamo.vc Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
On this episode of Dynamo Faithful, the lads review the match vs NE, chat about the transfer rumors, then preview the match vs LAG, and talk best duos.Be sure to follow us on Instagram and Twitter @Dynamo_Faithful and let us know what you think! We appreciate any feedback on how to improve the pod going forward, and please consider rating and reviewing us on your favorite podcast platform!Appearing on this episode are Chris Sinski & Krystopher Scroggins.Produced & Edited by Ian Gregory-GraffSocial Media & Design by Zacj BellotMusic from Pixabay:Intro/Outro Song: Indie Folk (King Around Here) by Alex Grohl
Welcome back to the Bayou City Soccer Podcast! This week, Manny and Joey join the guys to break down Houston's massive 2-0 road win over the New England Revolution. The Dynamo delivered another clean sheet, with Guilherme and Agustín Resch finding the net as Houston continues to build confidence and climb the MLS standings. The guys also preview what should be a huge celebration against the LA Galaxy as Houston honors the 20th anniversary of the 2006 championship team. Can the Dynamo keep the party going, or will the Galaxy play spoiler? Plus, the Archive Kit is dropping midweek, we discuss the latest transfers, and check in on D2's current form.-All that and much more in this episode!-We invite you to follow us at Bayou City Soccer!-BayouCitySoccer.net-@BayouCitySoccer on Facebook, IG, and Twitter-Hit us up using our hashtag #AskBCS-Hosts:@rudysegura3 on Twitter@crisputallaz on Twitter-Again, BayouCitySoccer.net for everything!
Welcome to another Orange Talk episode of the Bayou City Soccer Podcast! This week, we're joined by Houston Dynamo general manager and club legend Pat Onstad to discuss a team riding a six-match unbeaten streak, including three straight wins and clean sheets. We talk about the additions of Duncan McGuire and Marcelo Saracchi, the development of young players, the legacy of the 2006 championship team, and what the Dynamo can accomplish down the stretch.-We invite you to follow us at Bayou City Soccer!-BayouCitySoccer.net-@BayouCitySoccer on Facebook, IG, and Twitter-Hit us up using our hashtag #AskBCS-Hosts: Dustyn Richardson-Again, BayouCitySoccer.net for everything.-We invite you to follow us at Bayou City Soccer!-BayouCitySoccer.net-@BayouCitySoccer on Facebook, IG, and Twitter-Hit us up using our hashtag #AskBCS-Hosts: Dustyn Richardson-Again, BayouCitySoccer.net for everything.
Welcome to another Orange Talk episode of the Bayou City Soccer Podcast! This week, we're joined by Houston Dynamo 2 head coach Jeremy Hurdle to discuss a team sitting second in the Western Conference with 45 points. We talk about the challenge of roster turnover, Pedro Cruz's loan to El Paso, the emergence of young academy players, and the development of Mattheo Dimareli and JJ Bell. Plus, Jeremy looks at the addition of Victor André, the balance between winning and developing players, and Dynamo 2's push toward the playoffs.-We invite you to follow us at Bayou City Soccer!-BayouCitySoccer.net-@BayouCitySoccer on Facebook, IG, and Twitter-Hit us up using our hashtag #AskBCS-Hosts: Dustyn Richardson & Rudy Segura-Again, BayouCitySoccer.net for everything.
Selten ist der 1. FC Nürnberg so souverän in eine Saison gestartet wie am Sonntag gegen Dynamo Dresden. Nach dem 3:0 ist es einigermaßen kompliziert, das berühmte Haar in der Suppe zu finden, aber Sebastian Gloser, Uli Digmayer und Fadi Keblawi haben in dieser Disziplin ja eine gewisse Meisterschaft entwickelt. Wirklich geschimpft wird in der neuen und von der Sparkasse Nürnberg präsentierten Folge zwar auch, aber selten mit und über den 1. FC Nürnberg. Stattdessen stellt sich da die Frage, wie man Mohamed Ali Zoma davon überzeugen kann, auch über das Ende der Transferphase hinaus beim Club zu bleiben. Angebote dürfte es nicht erst seit seinen drei Toren vom Wochenende reichlich geben für den Angreifer. Wird er tatsächlich noch verkauft, dürfte das die Hoffnungen auf eine richtig erfolgreiche Zweitligasaison nicht nur ein wenig schmälern. Zu sehr wirkt der Ansatz von Miro Klose in diesem Fußballjahr auf Zoma zugeschnitten. Gegen Dresden konnten aber auch (fast) alle anderen überzeugen. Ansonsten wird erstmals zurückgerudert und irgendwie Österreich eingemeindet. Weitere Themen: das Brückenfestival, Telekommunikationskonzerne und agenturbetriebene Podcasts aus Fürth.
Clean sheets! Mark and OSG are joined by The Surge's Cook to chat Houston's win at Sporting KC, potential impact of Saracchi's signing, preview the trip to New England, and more. Timestamps: 00:00 Intro 02:33 Review: Sporting Kansas City 19:18 Houston climb standings and go for top of the West 25:22 Marcelo Saracchi signing confirmed by Cesar Luis Merlo 38:48 Preview: New England Revolution 53:40 Larry Berg announced as new MLS Commissioner 01:03:43 Do the Dynamo miss Leagues Cup? 01:09:43 Tailgate Announcements! Credits: ⬢ Noodle Time is hosted by Mark Segovia and OSG! ⬢ Today's guest is George from The Surge Supporters Group. ⬢ Intro/Outro music by Matt Houston. | Starfox - Armada [Matt Houston Remix] ⬢ Check out all of our content at DynamicFoxtrot.com. ⬢ Support Foxtrot Media on Ko-fi.com/DynamicFoxtrot. ⬢ Follow the fox on Twitter (@DynamicFoxtrot), Instagram (@dynamicfoxtrot), and Bluesky (@DynamicFoxtrot). ⬢ Subscribe to Foxtrot TV on YouTube! ⬢ Thumbnail photo provided by Houston Dynamo FC. Learn more about your ad choices. Visit megaphone.fm/adchoices
Welcome back to the Bayou City Soccer Podcast! This week, the guys are joined by Madelyn to recap the Houston Dynamo's road win over Sporting KC. A red card just before halftime tilted the match in Houston's favor, and the Dynamo found a way to close it out and secure three points. Guilherme continues his strong season, while Ezequiel Ponce is still searching for a goal. The crew also previews the matchup with the New England Revolution, including Brooklyn Raines facing his former club and the challenge of containing midfield maestro Carles Gil.-All that and much more in this episode!-We invite you to follow us at Bayou City Soccer!-BayouCitySoccer.net-@BayouCitySoccer on Facebook, IG, and Twitter-Hit us up using our hashtag #AskBCS-Hosts:@rudysegura3 on Twitter@crisputallaz on Twitter-Again, BayouCitySoccer.net for everything!
Hour 1 of Soccer Matters with Glenn Davis! Game on for 90 - are fans more interested in individuals players over team? Glenn touches on transfer market, new MLS Commissioner, + USMNT retaining Mauricio Pochettino, German Moores of the Houston Chronicle joins to talk World Cup + new Dynamo signing.
In this episode, Manna founder Bobby Healy explains how his company designs and operates autonomous electric delivery drones, achieving over 300,000 flights and 97% availability in harsh Irish weather by running operations like a low-cost airline focused on high utilization and efficiency. He discusses why a recent U.S. policy shift and forthcoming Part 108 regulations prompted Manna to go all-in on America, starting with a citywide drone delivery mesh in Tulsa, Oklahoma, chosen for its suburban density, aerospace ecosystem, and pro-drone stance. Bobby outlines why he sees players like Wing, Zipline, and Amazon more as fellow builders in a massive, non–winner-take-all market, and why partnering with aggregators such as DoorDash and Uber Eats is the most efficient go-to-market path. He also covers Manna's recent $50 million fundraise to scale manufacturing, operations, and R&D, and shares his 10-year vision in which drone delivery becomes the dominant, far cheaper, and faster last-mile solution for tens of millions of U.S. suburban homes. Highlights from their conversation include: Introducing Bobby Healy and Manna Overview (0:29) Making Drone Delivery Work Commercially Like a Low-Cost Airline (1:34) Why Now Is the Time for Manna To Enter the U.S. Market (3:28) Why Tulsa, Oklahoma Is Manna's U.S. Launch City (6:19) Competing With Wing, Zipline, Amazon, and Other Drone Players (8:04) Role of DoorDash, Uber Eats, and Aggregators in Manna's Strategy (11:24) Operational Playbook for Launching New Drone Cities (15:23) How Manna Will Use Its Recent $50 Million Fundraise (17:51) Ten-Year Vision for Drone-First Last-Mile Delivery in the U.S. (19:19) Final Thoughts and Takeaways (21:18) Dynamo Ventures is a venture firm backing founders upgrading the physical economy. As intelligence moves into critical infrastructure and technology collides with physics, industry is entering a new era of transformation - the industrial renaissance. Born from the dirt and grit of supply chains and shaped by operations, not spreadsheets, Dynamo focuses on the complex realities of building in the real world. We invest in companies transforming infrastructure, manufacturing, logistics, transportation, and the systems that power global commerce. Dynamo works closely with founders who combine ambition with a bias to action, bringing a builder mindset to venture capital through deep operational insight, systematic pressure-testing and hands-on partnership. Our purpose is simple: to back the relentless shaping the industrial renaissance. Learn more at www.dynamo.vc. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
On this episode of Dynamo Faithful, the lads review the match vs SKC, chat about the transfer rumors, Leagues Cup, & MLS Commissioner, then preview the match vs the Revs, and talk best newly released kits.Be sure to follow us on Instagram and Twitter @Dynamo_Faithful and let us know what you think! We appreciate any feedback on how to improve the pod going forward, and please consider rating and reviewing us on your favorite podcast platform!Appearing on this episode are Chris Sinski, Manny Farciert, & Kyle McGuire.Produced & Edited by Ian Gregory-GraffSocial Media & Design by Zacj BellotMusic from Pixabay:Intro/Outro Song: Indie Folk (King Around Here) by Alex Grohl
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
Chopped! Mark, Selvin, and yours truly chat Houston's demolition of Austin FC, Mati Bogusz showing up, Guilherme and Pedro Cruz featured in MLS All-Star Week, preview SKC, and more. Timestamps: 00:00 Intro 00:50 Review: Austin FC 10:30 Resurgence from Bogusz and Guilherme on a roll 20:40 HH's role coming back to Houston 33:48 Guilherme plays in 2026 MLS All-Star Game, Pedro Cruz loses Goalie Warz title 42:30 Preview: Sporting Kansas City 01:01:22 Dynamo reportedly going for Celtic, Boca Juniors wingback Marcelo Saracchi 01:10:40 Closing Credits: ⬢ Noodle Time is hosted by Mark Segovia, Selvin Garcia, and yours truly Andrés Naranjo! ⬢ Intro/Outro music by Matt Houston. | Starfox - Armada [Matt Houston Remix] ⬢ Check out all of our content at DynamicFoxtrot.com. ⬢ Support Foxtrot Media on Ko-fi.com/DynamicFoxtrot. ⬢ Follow the fox on Twitter (@DynamicFoxtrot), Instagram (@dynamicfoxtrot), and Bluesky (@DynamicFoxtrot). ⬢ Subscribe to Foxtrot TV on YouTube! ⬢ Thumbnail photo provided by Manuel Gonzalez - Foxtrot Media. Learn more about your ad choices. Visit megaphone.fm/adchoices
Als letzten Verein sprechen wir über Dynamo Dresden. Anne Vidal über den Wiederaufstieg, Beständigkeit und die Arbeit von Fanhilfen.
Hour 2 of Soccer Matters with Glenn Davis! The "Golden Age' of soccer development with Tom Byer, have you missed development years? Transfer window and "Italian" crazy? Maldini resigns Dynamo / Dash recap.
In this episode, Madelyn O'Farrell talks with Freight Hero Co-Founder and CEO, Andre Martins, about how his company uses AI-powered operations as a service to automate the highly manual world of freight brokerage. Andre explains the broker's cradle-to-grave responsibilities, why traditional software often fails in an exception-heavy, relationship-driven industry, and how Freight Hero instead “does the work” with an AI agent (Robin) plus a human-in-the-loop team. They discuss real-world results such as brokers growing revenue without adding headcount, surviving margin compression during the “Great Freight Recession,” and reallocating staff from low-value tracking tasks to higher-value sales and customer service. Andre also shares details on Freight Hero's recent seed round, their roadmap into billing and carrier sales, and his advice for brokers who are skeptical about adopting AI but need a low-friction way to modernize their operations. Highlights from their conversation include: Meet Andre Martins And Freight Hero Overview (0:29) Andre's Previous Startup And Path into Freight Brokerage (1:18) What Freight Brokers Actually Do Day to Day (3:27) Why Freight Hero Sells Outcomes, not Software (8:11) How Robin Learns And Operates Across 40,000+ Loads (13:19) Human-in-the-Loop Stories And Handling Driver Exceptions (17:23) Customer Results, Cost Pressure, And Great Freight Recession (18:31) Growing Revenue 80% With No New Headcount (21:53) Seed Round Announcement And Why Investors Bought In (22:12) Roadmap Beyond Track And Trace To Full Back Office (23:33) Vision For Brokerage Of the Future And Role Of Relationships (27:15) Lessons Learned And Advice For AI-Skeptical Brokers (28:01) Closing Thoughts And Episode Wrap-Up (30:56) Dynamo Ventures is a venture firm backing founders upgrading the physical economy. As intelligence moves into critical infrastructure and technology collides with physics, industry is entering a new era of transformation - the industrial renaissance. Born from the dirt and grit of supply chains and shaped by operations, not spreadsheets, Dynamo focuses on the complex realities of building in the real world. We invest in companies transforming infrastructure, manufacturing, logistics, transportation, and the systems that power global commerce. Dynamo works closely with founders who combine ambition with a bias to action, bringing a builder mindset to venture capital through deep operational insight, systematic pressure-testing and hands-on partnership. Our purpose is simple: to back the relentless shaping the industrial renaissance. Learn more at www.dynamo.vc Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
On this episode of Dynamo Faithful, the lads review the matches vs DC & Austin, chat about the sponsor, allstar, & rumors, preview the match vs SKC, and think about free agent sadness.Be sure to follow us on Instagram and Twitter @Dynamo_Faithful and let us know what you think! We appreciate any feedback on how to improve the pod going forward, and please consider rating and reviewing us on your favorite podcast platform!Appearing on this episode are Chris Sinski, Jake Berry, & Kyle McGuire.Produced & Edited by Ian Gregory-GraffSocial Media & Design by Zacj BellotMusic from Pixabay:Intro/Outro Song: Indie Folk (King Around Here) by Alex Grohl
Welcome back to the Bayou City Soccer Podcast! This week, the guys are joined by Madelyn to recap the Houston Dynamo's matches against D.C. United and Austin FC. The crew breaks down Houston's frustrating draw with D.C. United that felt more like a loss before celebrating a dominant 3-0 win over Texas rival Austin FC. With the second half of the season underway, the Dynamo look healthy and back in form. They also discuss Guilherme's MLS All-Star selection after a standout debut season that has him firmly in the Newcomer of the Year conversation. Looking ahead, they preview Houston's road match against Sporting Kansas City before wrapping up with a look at Houston Dynamo 2 and the club's recent form.-All that and much more in this episode!-We invite you to follow us at Bayou City Soccer!-BayouCitySoccer.net-@BayouCitySoccer on Facebook, IG, and Twitter-Hit us up using our hashtag #AskBCS-Hosts:@rudysegura3 on Twitter@crisputallaz on Twitter-Again, BayouCitySoccer.net for everything!
From pregame to postgame, Selvin Garcia, OSG, and German Benitez have you covered in Houston's dominant victory over Austin FC. Timestamps: 00:00 Preview at the Supporters Lot 14:54 First Half Highlights and Halftime Reactions 22:18 Second Half Highlights 23:36 Ben Olsen - press conference 38:28 Guilherme - press conference 50:30 Jonathan Bond - mix zone 51:59 Erik Sviatchenko - mix zone 53:23 Postgame Reactions Credits: ⬢ Today's Foxtrot Gameday was hosted by Selvin Garcia, OSG, and German Benitez! ⬢ Check out all of our content at DynamicFoxtrot.com. ⬢ Support Foxtrot Media on Ko-fi.com/DynamicFoxtrot. ⬢ Follow the fox on Twitter (@DynamicFoxtrot), Instagram (@dynamicfoxtrot), and Bluesky (@DynamicFoxtrot). ⬢ Subscribe to Foxtrot TV on YouTube! ⬢ Game highlights and press conference footage provided by Houston Dynamo FC. ⬢ Thumbnail photo provided by Manuel Gonzalez - Foxtrot Media. Learn more about your ad choices. Visit megaphone.fm/adchoices
Mayor Whitmire is giving away 10,000 free tickets to Dynamo/Dash games --> Ticket InfoJohn Toomey still visits his BBQ place everyday and it's one of the best in Houston65% of us say we talk out loud to ourselves at least once a day while at work
From pregame to postgame, Selvin Garcia and German Benitez are back in The Shell to cover the Dynamo's return to MLS play after the World Cup break. Timestamps: 00:00 Preview at Shell Energy Stadium 03:44 First Half Highlights and Halftime Reactions 08:19 Second Half Highlights 11:08 Ben Olsen - press conference 16:47 Duncan McGuire - press conference 20:06 Jonathan Bond - mix zone 22:36 Nathan Ordaz - mix zone (English/Español) 26:27 Postgame Reactions Credits: ⬢ Today's Foxtrot Gameday was hosted by Selvin Garcia and German Benitez! ⬢ Check out all of our content at DynamicFoxtrot.com. ⬢ Support Foxtrot Media on Ko-fi.com/DynamicFoxtrot. ⬢ Follow the fox on Twitter (@DynamicFoxtrot), Instagram (@dynamicfoxtrot), and Bluesky (@DynamicFoxtrot). ⬢ Subscribe to Foxtrot TV on YouTube! ⬢ Game highlights and press conference footage provided by Houston Dynamo FC. ⬢ Thumbnail photo provided by Diego Herrera - Foxtrot Media. Learn more about your ad choices. Visit megaphone.fm/adchoices
Welcome back to the BCS: Final Whistle! We break down the Dynamo's performance against DC United, key moments, standout players, Ben Olsen's tactics, and what this result means moving forward. Plus, we react to your biggest fan takes.-All that and much more in this episode!-We invite you to follow us at Bayou City Soccer!-BayouCitySoccer.net-@BayouCitySoccer on Facebook, IG, and Twitter-Hit us up using our hashtag #AskBCS-Again, BayouCitySoccer.net for everything!
In this episode with The New Stack Agents, Frederic Lardinois, NVIDIA's Joey Conway says advances in AI over the past year have dramatically improved the capabilities of local models, making them practical for enterprise and personal use alongside frontier cloud models. Rather than replacing large models, Conway envisions a “system of models” where specialized local models handle routine, cost-sensitive, or privacy-focused tasks, while larger frontier models tackle more complex reasoning. He explains that organizations can fine-tune smaller open models using domain-specific data, creating expert AI agents that reflect the specialized roles found within businesses. NVIDIA supports this ecosystem through open models, training tools, and software such as NeMo, Dynamo, and Nemotron. Conway also highlights the growing importance of agentic harnesses, which give AI models access to tools, memory, and iterative workflows, significantly improving performance and reducing costs. Looking ahead, he expects AI orchestration to become increasingly important, with intelligent routing systems selecting the right model for each task based on complexity, cost, latency, and data governance requirements, enabling enterprises to balance performance, security, and efficiency. Learn more from The New Stack around NVIDIA's latest updates in AI: Palantir and Nvidia want to change who owns government AI Nvidia's best model is now live Join our community of newsletter subscribers to stay on top of the news and at the top of your game.
Hour 2 of Soccer Matters with Glenn Davis, this hour Glenn was joined by soccer journalist Victor AraizaThey discussed their top 10 World Cup moments, the Houston Dynamo making an embarrassing sponsorship decision, and more
In this episode, Scott Friedman, VP of Government Affairs at Altana, joins Madelyn to trace his journey from crafting trade enforcement policy in government to building the technology that underpins it. He explains how trade has shifted from an era of assumed free trade to being a deliberate tool of state power, highlighting rapid changes in U.S. policy, the gap between policymakers' ambitions and what regulators and companies can practically execute, and the crucial role of modern data and technology. Scott outlines Altana's vision for a trusted, transparent global trade network, including product passports as a “global entry for goods” and a federated data architecture that enables collaboration without sacrificing privacy. He contrasts Western transparency-driven systems with China's state-directed, opaque but highly efficient digital trade ecosystem, and explores evolving U.S.–EU alignment on customs, traceability, and digital infrastructure. The key takeaway: full end-to-end traceability is fast becoming the baseline expectation, and companies that lean into data-driven transparency now will be far better positioned in an increasingly complex enforcement environment. Highlights from their conversation include: Scott's Journey from Government to Trade Tech (0:41) America Founded on a Customs Dispute and Boston Tea Party (2:47) How Today's Trade Enforcement Differs from Past Eras (4:54) Gap between Policymakers, Regulators, and Global Trade (7:46) Altana Overview and Vision for Trusted Global Trade (11:22) Product Passports as Global Entry for Goods (14:08) Why Altana Uses a Federated Network Model (19:06) China's State-Directed Supply Chain Architecture (22:19) U.S. and EU Alignment on Customs, Data, and Traceability (29:29) What End-to-End Traceability Now Means for Importers (34:37) Key Takeaways and Episode Wrap-Up (39:14) Dynamo Ventures is a venture firm backing founders upgrading the physical economy. As intelligence moves into critical infrastructure and technology collides with physics, industry is entering a new era of transformation - the industrial renaissance. Born from the dirt and grit of supply chains and shaped by operations, not spreadsheets, Dynamo focuses on the complex realities of building in the real world. We invest in companies transforming infrastructure, manufacturing, logistics, transportation, and the systems that power global commerce. Dynamo works closely with founders who combine ambition with a bias to action, bringing a builder mindset to venture capital through deep operational insight, systematic pressure-testing and hands-on partnership. Our purpose is simple: to back the relentless shaping the industrial renaissance. Learn more at www.dynamo.vc. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
Investor Fuel Real Estate Investing Mastermind - Audio Version
In this episode, Nicholas Battaglia shares insights into private lending for real estate, focusing on luxury ground-up construction and innovative financing solutions like 100% LTC. We explore market trends, challenges with appraisals, and strategies for scaling a real estate finance business. Professional Real Estate Investors - How we can help you: Investor Fuel Mastermind: Learn more about the Investor Fuel Mastermind, including 100% deal financing, massive discounts from vendors and sponsors you're already using, our world class community of over 150 members, and SO much more here: http://www.investorfuel.com/apply Investor Machine Marketing Partnership: Are you looking for consistent, high quality lead generation? Investor Machine is America's #1 lead generation service professional investors. Investor Machine provides true 'white glove' support to help you build the perfect marketing plan, then we'll execute it for you…talking and working together on an ongoing basis to help you hit YOUR goals! Learn more here: http://www.investormachine.com Coaching with Mike Hambright: Interested in 1 on 1 coaching with Mike Hambright? Mike coaches entrepreneurs looking to level up, build coaching or service based businesses (Mike runs multiple 7 and 8 figure a year businesses), building a coaching program and more. Learn more here: https://investorfuel.com/coachingwithmike Attend a Vacation/Mastermind Retreat with Mike Hambright: Interested in joining a "mini-mastermind" with Mike and his private clients on an upcoming "Retreat", either at locations like Cabo San Lucas, Napa, Park City ski trip, Yellowstone, or even at Mike's East Texas "Big H Ranch"? Learn more here: http://www.investorfuel.com/retreat Property Insurance: Join the largest and most investor friendly property insurance provider in 2 minutes. Free to join, and insure all your flips and rentals within minutes! There is NO easier insurance provider on the planet (turn insurance on or off in 1 minute without talking to anyone!), and there's no 15-30% agent mark up through this platform! Register here: https://myinvestorinsurance.com/ New Real Estate Investors - How we can work together: Investor Fuel Club (Coaching and Deal Partner Community): Looking to kickstart your real estate investing career? Join our one of a kind Coaching Community, Investor Fuel Club, where you'll get trained by some of the best real estate investors in America, and partner with them on deals! You don't need $ for deals…we'll partner with you and hold your hand along the way! Learn More here: http://www.investorfuel.com/club —--------------------
On this episode of Dynamo Faithful, the lads review the World Cup, chat about the roster news, preview the matches vs DC & Austin, and think about future World Cup formats.Be sure to follow us on Instagram and Twitter @Dynamo_Faithful and let us know what you think! We appreciate any feedback on how to improve the pod going forward, and please consider rating and reviewing us on your favorite podcast platform!Appearing on this episode are Chris Sinski, Krystopher Scroggins, Jake Berry, & Kyle McGuire.Produced & Edited by Ian Gregory-GraffSocial Media & Design by Zacj BellotMusic from Pixabay:Intro/Outro Song: Indie Folk (King Around Here) by Alex Grohl
Welcome back to the Bayou City Soccer Podcast! Rudy is joined by Madelyn to break down the World Cup Final between Spain and Argentina, covering everything surrounding the match, the good, the bad, and the ugly. They also share their favorite soccer memories from an unforgettable tournament. Then, they shift their focus back to MLS as the Houston Dynamo return to regular-season action with two home matches at Shell Energy Stadium this week: a midweek clash against D.C. United followed by a Saturday showdown with Austin FC. The crew also discusses the Dynamo's leaked third kit from Adidas and debates whether the club and MLS did enough to capitalize on the world's biggest soccer tournament to attract new fans and grow the league's ecosystem.-All that and much more in this episode!-We invite you to follow us at Bayou City Soccer!-BayouCitySoccer.net-@BayouCitySoccer on Facebook, IG, and Twitter-Hit us up using our hashtag #AskBCS-Hosts:@rudysegura3 on Twitter@crisputallaz on Twitter-Again, BayouCitySoccer.net for everything!
Welcome to another Orange Talk episode of the Bayou City Soccer Podcast! This week, we're joined by Houston Dynamo head coach Ben Olsen to discuss the club's return to MLS action following the league break. We dive into the confidence the team built through strong performances against CD Olimpia and América de Cali, the emergence of young players like Matthew Arana and Pedro Cruz as they push for first-team minutes, and what the Dynamo hope to bring to the city over the final 20 matches of the regular season.-We invite you to follow us at Bayou City Soccer!-BayouCitySoccer.net-@BayouCitySoccer on Facebook, IG, and Twitter-Hit us up using our hashtag #AskBCS-Hosts: Dustyn Richardson-Again, BayouCitySoccer.net for everything.
ITL reacts to a new report from Bill Barnwell after conversations with NFL executives about player trade values quickly turned into a debate over Texans quarterback C.J. Stroud. The guys discuss why opinions around the league are mixed, what Stroud must prove to re-establish himself among the NFL's elite quarterbacks, and whether the criticism is warranted heading into the season. B Scott joins the show for Around The NFL, shares his thoughts on Stroud, and brings his three biggest questions for the Texans before training camp. Reggie continues his Texans Re-Watch series by revisiting Houston's matchup with the Broncos, while Lunch-Time Confessions delivers another round of laughs before Houston Dynamo head coach Ben Olsen joins the show to discuss the World Cup, the Dynamo, and soccer's continued growth in Houston. ITL also debates one of the greatest Houston sports arguments: Who was the better defensive player, Hakeem Olajuwon or J.J. Watt? Plus, the crew looks ahead to the Astros' trade deadline strategy, discusses what Dana Brown could do to improve the roster, and wraps up with Figgy's Mixtape featuring an HVAC thief on the loose, a salute to a longtime Payne & Pendergast producer, and a farewell to the Nissan Altima.
Reggie shares his biggest takeaways after revisiting Houston's matchup against Denver. The crew shares another round of confessions, stories, and laughs. Houston Dynamo head coach Ben Olsen joins the show to discuss the Dynamo, the World Cup, and soccer's continued growth in Houston.
Back to it! Selvin and yours truly are joined by JC from The Surge Supporters Group and get caught up on all things Houston Dynamo prior to our return to MLS play against DC United. Timestamps: 00:00 Intro 02:27 What a World Cup, huh? 10:22 Roster updates and Duncan McGuire traded from Orlando City 21:27 Dynamo reps in MLS All-Star Week. Guilherme snubbed! 33:42 Preview: DC United 45:20 Join The Surge Tailgate at the Orange Lot! 49:30 Dynamo 2 updates. Jeremy Hurdle named permanent Head Coach 56:03 Last thoughts and Closing Credits: ⬢ Noodle Time is hosted by Mark Segovia , Selvin Garcia, and yours truly Andrés Naranjo! ⬢ Today's guest is JC from The Surge! ⬢ Intro/Outro music by Matt Houston. | Starfox - Armada [Matt Houston Remix] ⬢ Check out all of our content at DynamicFoxtrot.com. ⬢ Support Foxtrot Media on Ko-fi.com/DynamicFoxtrot. ⬢ Follow the fox on Twitter (@DynamicFoxtrot), Instagram (@dynamicfoxtrot), and Bluesky (@DynamicFoxtrot). ⬢ Subscribe to Foxtrot TV on YouTube! ⬢ Thumbnail photo provided by Raphael Fernandez - Foxtrot Media. Learn more about your ad choices. Visit megaphone.fm/adchoices
From pregame to postgame, Selvin Garcia goes through Houston's last tune-up prior to next week's restart of the MLS season against Colombia's América de Cali. Timestamps: 00:00 Preview at Shell Energy Stadium 01:30 First Half Highlights and Halftime Reactions 02:46 Second Half Highlights 03:14 Ben Olsen - mix zone 09:47 Artur - mix zone (Español) 13:56 Duncan McGuire - mix zone 15:17 Postgame Reactions Credits: ⬢ Today's Foxtrot Gameday was hosted by Selvin Garcia! ⬢ Check out all of our content at DynamicFoxtrot.com. ⬢ Support Foxtrot Media on Ko-fi.com/DynamicFoxtrot. ⬢ Follow the fox on Twitter (@DynamicFoxtrot), Instagram (@dynamicfoxtrot), and Bluesky (@DynamicFoxtrot). ⬢ Subscribe to Foxtrot TV on YouTube! ⬢ Game highlights and press conference footage provided by Houston Dynamo FC. ⬢ Thumbnail photo provided by Raphael Fernandez - Foxtrot Media. Learn more about your ad choices. Visit megaphone.fm/adchoices
From pregame to postgame, Selvin Garcia and German Benitez review Houston Dynamo taking on Honduras' CD Olimpia in one of their friendlies ahead of the restart of the 2026 regular season. Timestamps: 00:00 Preview at Shell Energy Stadium 03:15 First Half Highlights and Halftime Reactions 07:43 Second Half Highlights 08:23 Ben Olsen - mix zone 14:33 Mateusz Bogusz - mix zone 15:38 Artur - mix zone (Español) 17:34 Postgame Reactions Credits: ⬢ Today's Foxtrot Gameday was hosted by Selvin Garcia and German Benitez! ⬢ Check out all of our content at DynamicFoxtrot.com. ⬢ Support Foxtrot Media on Ko-fi.com/DynamicFoxtrot. ⬢ Follow the fox on Twitter (@DynamicFoxtrot), Instagram (@dynamicfoxtrot), and Bluesky (@DynamicFoxtrot). ⬢ Subscribe to Foxtrot TV on YouTube! ⬢ Game highlights and press conference footage provided by Houston Dynamo FC. ⬢ Thumbnail photo provided by Raphael Fernandez - Foxtrot Media. Learn more about your ad choices. Visit megaphone.fm/adchoices
In this episode, energy policy expert Erik Olson joins Madelyn O'Farrell to unpack how surging electricity demand from AI and data centers is colliding with an aging, slow-moving U.S. grid and regulatory system. They discuss why today's load growth is comparable to the 1970s air conditioning boom, the challenges utilities face in integrating massive, fast-moving data center projects, and the financial risks of overbuilding grid infrastructure that may never be fully used. Erik explains how geography, water constraints, transmission limits, and state-by-state rules shape where data centers can actually go, and highlights creative solutions like data centers directly funding clean energy projects or paying to improve residential energy efficiency to offset their load. He closes with lessons from his time at the Department of Energy, arguing that policymakers need to prioritize simplicity, speed, and clear implementation over “perfect” policy design if they want to keep up with the pace of real-world energy infrastructure development. Highlights from their conversation include: Erik's Background in Energy Policy (0:28) Explaining the Generational Surge in Electricity Demand (1:34) Collision Between Fast Tech Companies and Slow Grid Institutions (1:50) How Massive Data Centers Strain Traditional Utility Planning (3:54) Risks of Overbuilding, Stranded Assets, and Who Pays (7:37) Geographic Hotspots, Constraints, and State-by-State Differences (8:27) Creative Workarounds Like Curtailment, Direct Funding, and Efficiency (11:18) Lessons From DOE on Policy, Implementation, and Speed in Government (15:12) Final Thoughts on Moving at the Speed of Business (20:51) Dynamo Ventures is a venture firm backing founders upgrading the physical economy. As intelligence moves into critical infrastructure and technology collides with physics, industry is entering a new era of transformation - the industrial renaissance. Born from the dirt and grit of supply chains and shaped by operations, not spreadsheets, Dynamo focuses on the complex realities of building in the real world. We invest in companies transforming infrastructure, manufacturing, logistics, transportation, and the systems that power global commerce. Dynamo works closely with founders who combine ambition with a bias to action, bringing a builder mindset to venture capital through deep operational insight, systematic pressure-testing and hands-on partnership. Our purpose is simple: to back the relentless shaping the industrial renaissance. Learn more at www.dynamo.vc. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
Welcome to another Orange Talk episode of the Bayou City Soccer Podcast! We're joined by Houston Dynamo 2 head coach Jeremy Hurdle to discuss the team's challenging stretch of 10 consecutive road matches and how the players and staff managed that demanding schedule. We also look at the team's recent run of form since returning home, the continued development of the roster, and the club's player pathway, from academy prospects making their Dynamo 2 debuts to players earning first-team opportunities and others moving to new clubs in search of more consistent playing time.-We invite you to follow us at Bayou City Soccer!-BayouCitySoccer.net-@BayouCitySoccer on Facebook, IG, and Twitter-Hit us up using our hashtag #AskBCS-Hosts: Dustyn Richardson-Again, BayouCitySoccer.net for everything.
Welcome to another Orange Talk episode of the Bayou City Soccer Podcast! We're joined by new Houston Dynamo striker Duncan McGuire to discuss his fresh start in Houston, what drew him to the club, and what Dynamo fans can expect from his game. We also talk about life away from soccer, the excitement of the World Cup in the U.S., and his goals for the rest of the MLS season.-We invite you to follow us at Bayou City Soccer!-BayouCitySoccer.net-@BayouCitySoccer on Facebook, IG, and Twitter-Hit us up using our hashtag #AskBCS-Hosts: Dustyn Richardson-Again, BayouCitySoccer.net for everything.
Welcome back to the Bayou City Soccer Podcast! Cris is joined by Madelyn to break down everything happening at the FIFA World Cup as the tournament reaches its thrilling Final Four. With the championship match set for this Sunday, they share their predictions on who will lift the trophy. The conversation then shifts back to Houston, where the Dynamo are preparing for the second half of the season with a series of summer friendlies. Cris and Madelyn also discuss the latest club news, including the arrival of Duncan McGuire and what his addition could mean for Houston's attack as the team gears up for the stretch run.-All that and much more in this episode!-We invite you to follow us at Bayou City Soccer!-BayouCitySoccer.net-@BayouCitySoccer on Facebook, IG, and Twitter-Hit us up using our hashtag #AskBCS-Hosts:@rudysegura3 on Twitter@crisputallaz on Twitter-Again, BayouCitySoccer.net for everything!
Episode: 1600 Henry Adams ponders the Virgin and the Dynamo. science, medieval architecture, Langley, 1900 Paris exhibition, Chartres cathedral.
In this episode, Madelyn O'Farrell talks with Mike Matson, Co-Founder and CEO of Birch Geothermal, about his journey from the US Navy and academia through oil and gas, carbon capture, and consulting at BCG to founding Birch. They unpack what Enhanced Geothermal Systems (EGS) are, why Birch is “rooted in speed,” and how geothermal can scale quickly to meet soaring power demand from data centers and AI infrastructure. Mike explains geothermal's rare bipartisan appeal, the strong market signal from Fervo's recent IPO, and how 80–90% of oil and gas subsurface skills transfer directly into geothermal. They also explore what must go right for geothermal to move from a rounding error to a meaningful share of the US power mix over the next decade, and where the biggest “picks and shovels” investment opportunities lie across drilling, subsurface design, monitoring, and modular power systems. Highlights from their conversation include: Mike's Journey from Navy Service to Academia and Teaching (0:41) Midlife Rowing Expedition and Move into Drilling Operations (2:55) Shift from Oil and Gas to Carbon Capture and Clean Energy (3:42) Joining BCG and Discovering a Thesis Around Geothermal (4:31) Why Geothermal Has Unique Bipartisan Support in Politics (6:08) Market Signal from Fervo's Landmark IPO and Debt Financing (10:36) Why the Mountain West Is Ground Zero for Early EGS Projects (13:55) How Oil and Gas Skills Transfer 80–90% into Geothermal Roles (16:27) Birch's Thesis: Scaling Geothermal without a Green Premium (19:55) AI, Data Centers, and the 10-Year Outlook for Geothermal in the US Mix (23:19) Picks and Shovels Opportunities Across Drilling, Subsurface, and ORC Systems (25:51) Final Thoughts and Takeaways (29:37) Dynamo Ventures is a venture firm backing founders upgrading the physical economy. As intelligence moves into critical infrastructure and technology collides with physics, industry is entering a new era of transformation - the industrial renaissance. Born from the dirt and grit of supply chains and shaped by operations, not spreadsheets, Dynamo focuses on the complex realities of building in the real world. We invest in companies transforming infrastructure, manufacturing, logistics, transportation, and the systems that power global commerce. Dynamo works closely with founders who combine ambition with a bias to action, bringing a builder mindset to venture capital through deep operational insight, systematic pressure-testing and hands-on partnership. Our purpose is simple: to back the relentless shaping the industrial renaissance. Learn more at www.dynamo.vc. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
Part two of Secrets, KSTP Channel 5's Bailey Hurley calls in to share her own review of Dez's album, What We Learned, The News Dez Refused to UseSee Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.
© 2020-2026 Ivy Podcast Discovery LLC
