Podcasts about Latency

Analysts Don Kellogg and Roger Entner share insights from the recent T-Mobile Capital Markets Day update event, covering changes in the retail store strategy, new AI translation features, and updated reporting metrics.00:00 Episode intro 00:24 T-Mobile Capital Markets Day event overview 01:39 Pursuing convergence through fiber and FWA 02:37 Live AI translation could be a game-changer 04:03 Telecom must transcend the “dumb pipe” 05:08 Latency is key going forward 06:01 How AI actually impacts user experience 08:20 Key reporting metrics are changing 10:28 Episode wrap-upTags: telecom, telecommunications, wireless, prepaid, postpaid, cellular phone, Don Kellogg, Roger Entner, AI, T-Mobile, retail, FWA, fiber, convergence, USI, translation, network, Verizon, AT&T, latency, Apple, Android, churn, Srini Gopalan

Vodafone Ireland has once again been recognised as Ireland's leading mobile network, winning Best Mobile Internet Performance 2025 and Best 5G Network 2025 at this year's nPerf Awards, marking an exceptional fourth consecutive year in the top position. These latest wins follow years of continuous network improvement, with enhanced customer experience seen throughout the country. This result builds on last year's achievement, where Vodafone was awarded Best Mobile Internet Performance and Best 5G Provider for 2024. The 2025 results reinforce Vodafone's continued leadership in reliability, speed, and customer experience. This accolade further strengthens Vodafone Ireland's position as the country's most trusted mobile network, having also been independently recognised by umlaut as Ireland's Best Mobile Network for an unparalleled ten consecutive years. According to nPerf's independent analysis, based on thousands of real-world tests carried out by mobile users across Ireland, Vodafone achieved a total score of 98 361nPoints, outperforming all competitors across critical measures of network quality. The company excelled in multiple categories and was the best-performing provider across four – Upload Speed, Latency, Web Browsing, and YouTube Streaming – showcasing reliable performance in the 5G sector. Vodafone continues to advance its nationwide mobile network with €100 million invested annually to expand coverage, boost capacity, and deliver best-in-class mobile experiences. This sustained investment is driving significant modernisation across the network, including accelerated 4G and 5G rollout, enhanced resilience, and technology upgrades designed to meet Ireland's growing demand for fast, reliable connectivity. This investment forms part of Vodafone Ireland's wider €500 million five?year network investment cycle, which is delivering extensive upgrades and new site developments nationwide. As part of this long-term programme, Vodafone is delivering network enhancements across the country — recent examples in the last number of weeks include Carlow Town, Clane, Greystones, Mountcollins in Co. Limerick and the Mount Falcon region in Co. Mayo. These reflect just a snapshot of the wider, ongoing investment aimed at boosting coverage, increasing capacity and strengthening reliability for customers wherever they live, work and travel. In parallel, Vodafone continues to support Ireland's growing data demand, with annual mobile data traffic rising sharply at 20% year-on-year, ensuring customers can rely on a resilient, future proof network wherever they live, work, and travel. Sheila Kavanagh, Director of Networks, Vodafone Ireland, said: "We are incredibly proud to secure the nPerf award for the fourth year in a row. This result reflects the dedication of our teams and the impact of our continued investment, including €100 million each year — to deliver a fast, reliable and resilient network for customers across Ireland. As data usage grows and customer expectations evolve, we remain focused on building a future proof network that supports homes, businesses and communities, from major cities to remote rural areas. Our ambition is simple: to provide an outstanding, high quality experience every single day." Recognising Vodafone's continued success in 2025 Sébastien de Rosbo, Managing Director nPerf, stated: "Vodafone Ireland once again achieved the highest overall performance across our 2025 benchmarks. Based on large-scale, real-world testing carried out by users nationwide, Vodafone delivered consistently strong results across key indicators including mobile internet performance and 5G experience. This fourth consecutive win reflects the strength and consistency of Vodafone's network for customers across Ireland." See more stories here.

From rewriting Google's search stack in the early 2000s to reviving sparse trillion-parameter models and co-designing TPUs with frontier ML research, Jeff Dean has quietly shaped nearly every layer of the modern AI stack. As Chief AI Scientist at Google and a driving force behind Gemini, Jeff has lived through multiple scaling revolutions from CPUs and sharded indices to multimodal models that reason across text, video, and code.Jeff joins us to unpack what it really means to “own the Pareto frontier,” why distillation is the engine behind every Flash model breakthrough, how energy (in picojoules) not FLOPs is becoming the true bottleneck, what it was like leading the charge to unify all of Google's AI teams, and why the next leap won't come from bigger context windows alone, but from systems that give the illusion of attending to trillions of tokens.We discuss:* Jeff's early neural net thesis in 1990: parallel training before it was cool, why he believed scaling would win decades early, and the “bigger model, more data, better results” mantra that held for 15 years* The evolution of Google Search: sharding, moving the entire index into memory in 2001, softening query semantics pre-LLMs, and why retrieval pipelines already resemble modern LLM systems* Pareto frontier strategy: why you need both frontier “Pro” models and low-latency “Flash” models, and how distillation lets smaller models surpass prior generations* Distillation deep dive: ensembles → compression → logits as soft supervision, and why you need the biggest model to make the smallest one good* Latency as a first-class objective: why 10–50x lower latency changes UX entirely, and how future reasoning workloads will demand 10,000 tokens/sec* Energy-based thinking: picojoules per bit, why moving data costs 1000x more than a multiply, batching through the lens of energy, and speculative decoding as amortization* TPU co-design: predicting ML workloads 2–6 years out, speculative hardware features, precision reduction, sparsity, and the constant feedback loop between model architecture and silicon* Sparse models and “outrageously large” networks: trillions of parameters with 1–5% activation, and why sparsity was always the right abstraction* Unified vs. specialized models: abandoning symbolic systems, why general multimodal models tend to dominate vertical silos, and when vertical fine-tuning still makes sense* Long context and the illusion of scale: beyond needle-in-a-haystack benchmarks toward systems that narrow trillions of tokens to 117 relevant documents* Personalized AI: attending to your emails, photos, and documents (with permission), and why retrieval + reasoning will unlock deeply personal assistants* Coding agents: 50 AI interns, crisp specifications as a new core skill, and how ultra-low latency will reshape human–agent collaboration* Why ideas still matter: transformers, sparsity, RL, hardware, systems — scaling wasn't blind; the pieces had to multiply togetherShow Notes:* Gemma 3 Paper* Gemma 3* Gemini 2.5 Report* Jeff Dean's “Software Engineering Advice fromBuilding Large-Scale Distributed Systems” Presentation (with Back of the Envelope Calculations)* Latency Numbers Every Programmer Should Know by Jeff Dean* The Jeff Dean Facts* Jeff Dean Google Bio* Jeff Dean on “Important AI Trends” @Stanford AI Club* Jeff Dean & Noam Shazeer — 25 years at Google (Dwarkesh)—Jeff Dean* LinkedIn: https://www.linkedin.com/in/jeff-dean-8b212555* X: https://x.com/jeffdeanGoogle* https://google.com* https://deepmind.googleFull Video EpisodeTimestamps00:00:04 — Introduction: Alessio & Swyx welcome Jeff Dean, chief AI scientist at Google, to the Latent Space podcast00:00:30 — Owning the Pareto Frontier & balancing frontier vs low-latency models00:01:31 — Frontier models vs Flash models + role of distillation00:03:52 — History of distillation and its original motivation00:05:09 — Distillation's role in modern model scaling00:07:02 — Model hierarchy (Flash, Pro, Ultra) and distillation sources00:07:46 — Flash model economics & wide deployment00:08:10 — Latency importance for complex tasks00:09:19 — Saturation of some tasks and future frontier tasks00:11:26 — On benchmarks, public vs internal00:12:53 — Example long-context benchmarks & limitations00:15:01 — Long-context goals: attending to trillions of tokens00:16:26 — Realistic use cases beyond pure language00:18:04 — Multimodal reasoning and non-text modalities00:19:05 — Importance of vision & motion modalities00:20:11 — Video understanding example (extracting structured info)00:20:47 — Search ranking analogy for LLM retrieval00:23:08 — LLM representations vs keyword search00:24:06 — Early Google search evolution & in-memory index00:26:47 — Design principles for scalable systems00:28:55 — Real-time index updates & recrawl strategies00:30:06 — Classic “Latency numbers every programmer should know”00:32:09 — Cost of memory vs compute and energy emphasis00:34:33 — TPUs & hardware trade-offs for serving models00:35:57 — TPU design decisions & co-design with ML00:38:06 — Adapting model architecture to hardware00:39:50 — Alternatives: energy-based models, speculative decoding00:42:21 — Open research directions: complex workflows, RL00:44:56 — Non-verifiable RL domains & model evaluation00:46:13 — Transition away from symbolic systems toward unified LLMs00:47:59 — Unified models vs specialized ones00:50:38 — Knowledge vs reasoning & retrieval + reasoning00:52:24 — Vertical model specialization & modules00:55:21 — Token count considerations for vertical domains00:56:09 — Low resource languages & contextual learning00:59:22 — Origins: Dean's early neural network work01:10:07 — AI for coding & human–model interaction styles01:15:52 — Importance of crisp specification for coding agents01:19:23 — Prediction: personalized models & state retrieval01:22:36 — Token-per-second targets (10k+) and reasoning throughput01:23:20 — Episode conclusion and thanksTranscriptAlessio Fanelli [00:00:04]: Hey everyone, welcome to the Latent Space podcast. This is Alessio, founder of Kernel Labs, and I'm joined by Swyx, editor of Latent Space. Shawn Wang [00:00:11]: Hello, hello. We're here in the studio with Jeff Dean, chief AI scientist at Google. Welcome. Thanks for having me. It's a bit surreal to have you in the studio. I've watched so many of your talks, and obviously your career has been super legendary. So, I mean, congrats. I think the first thing must be said, congrats on owning the Pareto Frontier.Jeff Dean [00:00:30]: Thank you, thank you. Pareto Frontiers are good. It's good to be out there.Shawn Wang [00:00:34]: Yeah, I mean, I think it's a combination of both. You have to own the Pareto Frontier. You have to have like frontier capability, but also efficiency, and then offer that range of models that people like to use. And, you know, some part of this was started because of your hardware work. Some part of that is your model work, and I'm sure there's lots of secret sauce that you guys have worked on cumulatively. But, like, it's really impressive to see it all come together in, like, this slittily advanced.Jeff Dean [00:01:04]: Yeah, yeah. I mean, I think, as you say, it's not just one thing. It's like a whole bunch of things up and down the stack. And, you know, all of those really combine to help make UNOS able to make highly capable large models, as well as, you know, software techniques to get those large model capabilities into much smaller, lighter weight models that are, you know, much more cost effective and lower latency, but still, you know, quite capable for their size. Yeah.Alessio Fanelli [00:01:31]: How much pressure do you have on, like, having the lower bound of the Pareto Frontier, too? I think, like, the new labs are always trying to push the top performance frontier because they need to raise more money and all of that. And you guys have billions of users. And I think initially when you worked on the CPU, you were thinking about, you know, if everybody that used Google, we use the voice model for, like, three minutes a day, they were like, you need to double your CPU number. Like, what's that discussion today at Google? Like, how do you prioritize frontier versus, like, we have to do this? How do we actually need to deploy it if we build it?Jeff Dean [00:02:03]: Yeah, I mean, I think we always want to have models that are at the frontier or pushing the frontier because I think that's where you see what capabilities now exist that didn't exist at the sort of slightly less capable last year's version or last six months ago version. At the same time, you know, we know those are going to be really useful for a bunch of use cases, but they're going to be a bit slower and a bit more expensive than people might like for a bunch of other broader models. So I think what we want to do is always have kind of a highly capable sort of affordable model that enables a whole bunch of, you know, lower latency use cases. People can use them for agentic coding much more readily and then have the high-end, you know, frontier model that is really useful for, you know, deep reasoning, you know, solving really complicated math problems, those kinds of things. And it's not that. One or the other is useful. They're both useful. So I think we'd like to do both. And also, you know, through distillation, which is a key technique for making the smaller models more capable, you know, you have to have the frontier model in order to then distill it into your smaller model. So it's not like an either or choice. You sort of need that in order to actually get a highly capable, more modest size model. Yeah.Alessio Fanelli [00:03:24]: I mean, you and Jeffrey came up with the solution in 2014.Jeff Dean [00:03:28]: Don't forget, L'Oreal Vinyls as well. Yeah, yeah.Alessio Fanelli [00:03:30]: A long time ago. But like, I'm curious how you think about the cycle of these ideas, even like, you know, sparse models and, you know, how do you reevaluate them? How do you think about in the next generation of model, what is worth revisiting? Like, yeah, they're just kind of like, you know, you worked on so many ideas that end up being influential, but like in the moment, they might not feel that way necessarily. Yeah.Jeff Dean [00:03:52]: I mean, I think distillation was originally motivated because we were seeing that we had a very large image data set at the time, you know, 300 million images that we could train on. And we were seeing that if you create specialists for different subsets of those image categories, you know, this one's going to be really good at sort of mammals, and this one's going to be really good at sort of indoor room scenes or whatever, and you can cluster those categories and train on an enriched stream of data after you do pre-training on a much broader set of images. You get much better performance. If you then treat that whole set of maybe 50 models you've trained as a large ensemble, but that's not a very practical thing to serve, right? So distillation really came about from the idea of, okay, what if we want to actually serve that and train all these independent sort of expert models and then squish it into something that actually fits in a form factor that you can actually serve? And that's, you know, not that different from what we're doing today. You know, often today we're instead of having an ensemble of 50 models. We're having a much larger scale model that we then distill into a much smaller scale model.Shawn Wang [00:05:09]: Yeah. A part of me also wonders if distillation also has a story with the RL revolution. So let me maybe try to articulate what I mean by that, which is you can, RL basically spikes models in a certain part of the distribution. And then you have to sort of, well, you can spike models, but usually sometimes... It might be lossy in other areas and it's kind of like an uneven technique, but you can probably distill it back and you can, I think that the sort of general dream is to be able to advance capabilities without regressing on anything else. And I think like that, that whole capability merging without loss, I feel like it's like, you know, some part of that should be a distillation process, but I can't quite articulate it. I haven't seen much papers about it.Jeff Dean [00:06:01]: Yeah, I mean, I tend to think of one of the key advantages of distillation is that you can have a much smaller model and you can have a very large, you know, training data set and you can get utility out of making many passes over that data set because you're now getting the logits from the much larger model in order to sort of coax the right behavior out of the smaller model that you wouldn't otherwise get with just the hard labels. And so, you know, I think that's what we've observed. Is you can get, you know, very close to your largest model performance with distillation approaches. And that seems to be, you know, a nice sweet spot for a lot of people because it enables us to kind of, for multiple Gemini generations now, we've been able to make the sort of flash version of the next generation as good or even substantially better than the previous generations pro. And I think we're going to keep trying to do that because that seems like a good trend to follow.Shawn Wang [00:07:02]: So, Dara asked, so it was the original map was Flash Pro and Ultra. Are you just sitting on Ultra and distilling from that? Is that like the mother load?Jeff Dean [00:07:12]: I mean, we have a lot of different kinds of models. Some are internal ones that are not necessarily meant to be released or served. Some are, you know, our pro scale model and we can distill from that as well into our Flash scale model. So I think, you know, it's an important set of capabilities to have and also inference time scaling. It can also be a useful thing to improve the capabilities of the model.Shawn Wang [00:07:35]: And yeah, yeah, cool. Yeah. And obviously, I think the economy of Flash is what led to the total dominance. I think the latest number is like 50 trillion tokens. I don't know. I mean, obviously, it's changing every day.Jeff Dean [00:07:46]: Yeah, yeah. But, you know, by market share, hopefully up.Shawn Wang [00:07:50]: No, I mean, there's no I mean, there's just the economics wise, like because Flash is so economical, like you can use it for everything. Like it's in Gmail now. It's in YouTube. Like it's yeah. It's in everything.Jeff Dean [00:08:02]: We're using it more in our search products of various AI mode reviews.Shawn Wang [00:08:05]: Oh, my God. Flash past the AI mode. Oh, my God. Yeah, that's yeah, I didn't even think about that.Jeff Dean [00:08:10]: I mean, I think one of the things that is quite nice about the Flash model is not only is it more affordable, it's also a lower latency. And I think latency is actually a pretty important characteristic for these models because we're going to want models to do much more complicated things that are going to involve, you know, generating many more tokens from when you ask the model to do so. So, you know, if you're going to ask the model to do something until it actually finishes what you ask it to do, because you're going to ask now, not just write me a for loop, but like write me a whole software package to do X or Y or Z. And so having low latency systems that can do that seems really important. And Flash is one direction, one way of doing that. You know, obviously our hardware platforms enable a bunch of interesting aspects of our, you know, serving stack as well, like TPUs, the interconnect between. Chips on the TPUs is actually quite, quite high performance and quite amenable to, for example, long context kind of attention operations, you know, having sparse models with lots of experts. These kinds of things really, really matter a lot in terms of how do you make them servable at scale.Alessio Fanelli [00:09:19]: Yeah. Does it feel like there's some breaking point for like the proto Flash distillation, kind of like one generation delayed? I almost think about almost like the capability as a. In certain tasks, like the pro model today is a saturated, some sort of task. So next generation, that same task will be saturated at the Flash price point. And I think for most of the things that people use models for at some point, the Flash model in two generation will be able to do basically everything. And how do you make it economical to like keep pushing the pro frontier when a lot of the population will be okay with the Flash model? I'm curious how you think about that.Jeff Dean [00:09:59]: I mean, I think that's true. If your distribution of what people are asking people, the models to do is stationary, right? But I think what often happens is as the models become more capable, people ask them to do more, right? So, I mean, I think this happens in my own usage. Like I used to try our models a year ago for some sort of coding task, and it was okay at some simpler things, but wouldn't do work very well for more complicated things. And since then, we've improved dramatically on the more complicated coding tasks. And now I'll ask it to do much more complicated things. And I think that's true, not just of coding, but of, you know, now, you know, can you analyze all the, you know, renewable energy deployments in the world and give me a report on solar panel deployment or whatever. That's a very complicated, you know, more complicated task than people would have asked a year ago. And so you are going to want more capable models to push the frontier in the absence of what people ask the models to do. And that also then gives us. Insight into, okay, where does the, where do things break down? How can we improve the model in these, these particular areas, uh, in order to sort of, um, make the next generation even better.Alessio Fanelli [00:11:11]: Yeah. Are there any benchmarks or like test sets they use internally? Because it's almost like the same benchmarks get reported every time. And it's like, all right, it's like 99 instead of 97. Like, how do you have to keep pushing the team internally to it? Or like, this is what we're building towards. Yeah.Jeff Dean [00:11:26]: I mean, I think. Benchmarks, particularly external ones that are publicly available. Have their utility, but they often kind of have a lifespan of utility where they're introduced and maybe they're quite hard for current models. You know, I, I like to think of the best kinds of benchmarks are ones where the initial scores are like 10 to 20 or 30%, maybe, but not higher. And then you can sort of work on improving that capability for, uh, whatever it is, the benchmark is trying to assess and get it up to like 80, 90%, whatever. I, I think once it hits kind of 95% or something, you get very diminishing returns from really focusing on that benchmark, cuz it's sort of, it's either the case that you've now achieved that capability, or there's also the issue of leakage in public data or very related kind of data being, being in your training data. Um, so we have a bunch of held out internal benchmarks that we really look at where we know that wasn't represented in the training data at all. There are capabilities that we want the model to have. Um, yeah. Yeah. Um, that it doesn't have now, and then we can work on, you know, assessing, you know, how do we make the model better at these kinds of things? Is it, we need different kind of data to train on that's more specialized for this particular kind of task. Do we need, um, you know, a bunch of, uh, you know, architectural improvements or some sort of, uh, model capability improvements, you know, what would help make that better?Shawn Wang [00:12:53]: Is there, is there such an example that you, uh, a benchmark inspired in architectural improvement? Like, uh, I'm just kind of. Jumping on that because you just.Jeff Dean [00:13:02]: Uh, I mean, I think some of the long context capability of the, of the Gemini models that came, I guess, first in 1.5 really were about looking at, okay, we want to have, um, you know,Shawn Wang [00:13:15]: immediately everyone jumped to like completely green charts of like, everyone had, I was like, how did everyone crack this at the same time? Right. Yeah. Yeah.Jeff Dean [00:13:23]: I mean, I think, um, and once you're set, I mean, as you say that needed single needle and a half. Hey, stack benchmark is really saturated for at least context links up to 1, 2 and K or something. Don't actually have, you know, much larger than 1, 2 and 8 K these days or two or something. We're trying to push the frontier of 1 million or 2 million context, which is good because I think there are a lot of use cases where. Yeah. You know, putting a thousand pages of text or putting, you know, multiple hour long videos and the context and then actually being able to make use of that as useful. Try to, to explore the über graduation are fairly large. But the single needle in a haystack benchmark is sort of saturated. So you really want more complicated, sort of multi-needle or more realistic, take all this content and produce this kind of answer from a long context that sort of better assesses what it is people really want to do with long context. Which is not just, you know, can you tell me the product number for this particular thing?Shawn Wang [00:14:31]: Yeah, it's retrieval. It's retrieval within machine learning. It's interesting because I think the more meta level I'm trying to operate at here is you have a benchmark. You're like, okay, I see the architectural thing I need to do in order to go fix that. But should you do it? Because sometimes that's an inductive bias, basically. It's what Jason Wei, who used to work at Google, would say. Exactly the kind of thing. Yeah, you're going to win. Short term. Longer term, I don't know if that's going to scale. You might have to undo that.Jeff Dean [00:15:01]: I mean, I like to sort of not focus on exactly what solution we're going to derive, but what capability would you want? And I think we're very convinced that, you know, long context is useful, but it's way too short today. Right? Like, I think what you would really want is, can I attend to the internet while I answer my question? Right? But that's not going to happen. I think that's going to be solved by purely scaling the existing solutions, which are quadratic. So a million tokens kind of pushes what you can do. You're not going to do that to a trillion tokens, let alone, you know, a billion tokens, let alone a trillion. But I think if you could give the illusion that you can attend to trillions of tokens, that would be amazing. You'd find all kinds of uses for that. You would have attend to the internet. You could attend to the pixels of YouTube and the sort of deeper representations that we can find. You could attend to the form for a single video, but across many videos, you know, on a personal Gemini level, you could attend to all of your personal state with your permission. So like your emails, your photos, your docs, your plane tickets you have. I think that would be really, really useful. And the question is, how do you get algorithmic improvements and system level improvements that get you to something where you actually can attend to trillions of tokens? Right. In a meaningful way. Yeah.Shawn Wang [00:16:26]: But by the way, I think I did some math and it's like, if you spoke all day, every day for eight hours a day, you only generate a maximum of like a hundred K tokens, which like very comfortably fits.Jeff Dean [00:16:38]: Right. But if you then say, okay, I want to be able to understand everything people are putting on videos.Shawn Wang [00:16:46]: Well, also, I think that the classic example is you start going beyond language into like proteins and whatever else is extremely information dense. Yeah. Yeah.Jeff Dean [00:16:55]: I mean, I think one of the things about Gemini's multimodal aspects is we've always wanted it to be multimodal from the start. And so, you know, that sometimes to people means text and images and video sort of human-like and audio, audio, human-like modalities. But I think it's also really useful to have Gemini know about non-human modalities. Yeah. Like LIDAR sensor data from. Yes. Say, Waymo vehicles or. Like robots or, you know, various kinds of health modalities, x-rays and MRIs and imaging and genomics information. And I think there's probably hundreds of modalities of data where you'd like the model to be able to at least be exposed to the fact that this is an interesting modality and has certain meaning in the world. Where even if you haven't trained on all the LIDAR data or MRI data, you could have, because maybe that's not, you know, it doesn't make sense in terms of trade-offs of. You know, what you include in your main pre-training data mix, at least including a little bit of it is actually quite useful. Yeah. Because it sort of tempts the model that this is a thing.Shawn Wang [00:18:04]: Yeah. Do you believe, I mean, since we're on this topic and something I just get to ask you all the questions I always wanted to ask, which is fantastic. Like, are there some king modalities, like modalities that supersede all the other modalities? So a simple example was Vision can, on a pixel level, encode text. And DeepSeq had this DeepSeq CR paper that did that. Vision. And Vision has also been shown to maybe incorporate audio because you can do audio spectrograms and that's, that's also like a Vision capable thing. Like, so, so maybe Vision is just the king modality and like. Yeah.Jeff Dean [00:18:36]: I mean, Vision and Motion are quite important things, right? Motion. Well, like video as opposed to static images, because I mean, there's a reason evolution has evolved eyes like 23 independent ways, because it's such a useful capability for sensing the world around you, which is really what we want these models to be. So I think the only thing that we can be able to do is interpret the things we're seeing or the things we're paying attention to and then help us in using that information to do things. Yeah.Shawn Wang [00:19:05]: I think motion, you know, I still want to shout out, I think Gemini, still the only native video understanding model that's out there. So I use it for YouTube all the time. Nice.Jeff Dean [00:19:15]: Yeah. Yeah. I mean, it's actually, I think people kind of are not necessarily aware of what the Gemini models can actually do. Yeah. Like I have an example I've used in one of my talks. It had like, it was like a YouTube highlight video of 18 memorable sports moments across the last 20 years or something. So it has like Michael Jordan hitting some jump shot at the end of the finals and, you know, some soccer goals and things like that. And you can literally just give it the video and say, can you please make me a table of what all these different events are? What when the date is when they happened? And a short description. And so you get like now an 18 row table of that information extracted from the video, which is, you know, not something most people think of as like a turn video into sequel like table.Alessio Fanelli [00:20:11]: Has there been any discussion inside of Google of like, you mentioned tending to the whole internet, right? Google, it's almost built because a human cannot tend to the whole internet and you need some sort of ranking to find what you need. Yep. That ranking is like much different for an LLM because you can expect a person to look at maybe the first five, six links in a Google search versus for an LLM. Should you expect to have 20 links that are highly relevant? Like how do you internally figure out, you know, how do we build the AI mode that is like maybe like much broader search and span versus like the more human one? Yeah.Jeff Dean [00:20:47]: I mean, I think even pre-language model based work, you know, our ranking systems would be built to start. I mean, I think even pre-language model based work, you know, our ranking systems would be built to start. With a giant number of web pages in our index, many of them are not relevant. So you identify a subset of them that are relevant with very lightweight kinds of methods. You know, you're down to like 30,000 documents or something. And then you gradually refine that to apply more and more sophisticated algorithms and more and more sophisticated sort of signals of various kinds in order to get down to ultimately what you show, which is, you know, the final 10 results or, you know, 10 results plus. Other kinds of information. And I think an LLM based system is not going to be that dissimilar, right? You're going to attend to trillions of tokens, but you're going to want to identify, you know, what are the 30,000 ish documents that are with the, you know, maybe 30 million interesting tokens. And then how do you go from that into what are the 117 documents I really should be paying attention to in order to carry out the tasks that the user has asked? And I think, you know, you can imagine systems where you have, you know, a lot of highly parallel processing to identify those initial 30,000 candidates, maybe with very lightweight kinds of models. Then you have some system that sort of helps you narrow down from 30,000 to the 117 with maybe a little bit more sophisticated model or set of models. And then maybe the final model is the thing that looks. So the 117 things that might be your most capable model. So I think it has to, it's going to be some system like that, that is really enables you to give the illusion of attending to trillions of tokens. Sort of the way Google search gives you, you know, not the illusion, but you are searching the internet, but you're finding, you know, a very small subset of things that are, that are relevant.Shawn Wang [00:22:47]: Yeah. I often tell a lot of people that are not steeped in like Google search history that, well, you know, like Bert was. Like he was like basically immediately inside of Google search and that improves results a lot, right? Like I don't, I don't have any numbers off the top of my head, but like, I'm sure you guys, that's obviously the most important numbers to Google. Yeah.Jeff Dean [00:23:08]: I mean, I think going to an LLM based representation of text and words and so on enables you to get out of the explicit hard notion of, of particular words having to be on the page, but really getting at the notion of this topic of this page or this page. Paragraph is highly relevant to this query. Yeah.Shawn Wang [00:23:28]: I don't think people understand how much LLMs have taken over all these very high traffic system, very high traffic. Yeah. Like it's Google, it's YouTube. YouTube has this like semantics ID thing where it's just like every token or every item in the vocab is a YouTube video or something that predicts the video using a code book, which is absurd to me for YouTube size.Jeff Dean [00:23:50]: And then most recently GROK also for, for XAI, which is like, yeah. I mean, I'll call out even before LLMs were used extensively in search, we put a lot of emphasis on softening the notion of what the user actually entered into the query.Shawn Wang [00:24:06]: So do you have like a history of like, what's the progression? Oh yeah.Jeff Dean [00:24:09]: I mean, I actually gave a talk in, uh, I guess, uh, web search and data mining conference in 2009, uh, where we never actually published any papers about the origins of Google search, uh, sort of, but we went through sort of four or five or six. generations, four or five or six generations of, uh, redesigning of the search and retrieval system, uh, from about 1999 through 2004 or five. And that talk is really about that evolution. And one of the things that really happened in 2001 was we were sort of working to scale the system in multiple dimensions. So one is we wanted to make our index bigger, so we could retrieve from a larger index, which always helps your quality in general. Uh, because if you don't have the page in your index, you're going to not do well. Um, and then we also needed to scale our capacity because we were, our traffic was growing quite extensively. Um, and so we had, you know, a sharded system where you have more and more shards as the index grows, you have like 30 shards. And then if you want to double the index size, you make 60 shards so that you can bound the latency by which you respond for any particular user query. Um, and then as traffic grows, you add, you add more and more replicas of each of those. And so we eventually did the math that realized that in a data center where we had say 60 shards and, um, you know, 20 copies of each shard, we now had 1200 machines, uh, with disks. And we did the math and we're like, Hey, one copy of that index would actually fit in memory across 1200 machines. So in 2001, we introduced, uh, we put our entire index in memory and what that enabled from a quality perspective was amazing. Um, and so we had more and more replicas of each of those. Before you had to be really careful about, you know, how many different terms you looked at for a query, because every one of them would involve a disk seek on every one of the 60 shards. And so you, as you make your index bigger, that becomes even more inefficient. But once you have the whole index in memory, it's totally fine to have 50 terms you throw into the query from the user's original three or four word query, because now you can add synonyms like restaurant and restaurants and cafe and, uh, you know, things like that. Uh, bistro and all these things. And you can suddenly start, uh, sort of really, uh, getting at the meaning of the word as opposed to the exact semantic form the user typed in. And that was, you know, 2001, very much pre LLM, but really it was about softening the, the strict definition of what the user typed in order to get at the meaning.Alessio Fanelli [00:26:47]: What are like principles that you use to like design the systems, especially when you have, I mean, in 2001, the internet is like. Doubling, tripling every year in size is not like, uh, you know, and I think today you kind of see that with LLMs too, where like every year the jumps in size and like capabilities are just so big. Are there just any, you know, principles that you use to like, think about this? Yeah.Jeff Dean [00:27:08]: I mean, I think, uh, you know, first, whenever you're designing a system, you want to understand what are the sort of design parameters that are going to be most important in designing that, you know? So, you know, how many queries per second do you need to handle? How big is the internet? How big is the index you need to handle? How much data do you need to keep for every document in the index? How are you going to look at it when you retrieve things? Um, what happens if traffic were to double or triple, you know, will that system work well? And I think a good design principle is you're going to want to design a system so that the most important characteristics could scale by like factors of five or 10, but probably not beyond that because often what happens is if you design a system for X. And something suddenly becomes a hundred X, that would enable a very different point in the design space that would not make sense at X. But all of a sudden at a hundred X makes total sense. So like going from a disk space index to a in memory index makes a lot of sense once you have enough traffic, because now you have enough replicas of the sort of state on disk that those machines now actually can hold, uh, you know, a full copy of the, uh, index and memory. Yeah. And that all of a sudden enabled. A completely different design that wouldn't have been practical before. Yeah. Um, so I'm, I'm a big fan of thinking through designs in your head, just kind of playing with the design space a little before you actually do a lot of writing of code. But, you know, as you said, in the early days of Google, we were growing the index, uh, quite extensively. We were growing the update rate of the index. So the update rate actually is the parameter that changed the most. Surprising. So it used to be once a month.Shawn Wang [00:28:55]: Yeah.Jeff Dean [00:28:56]: And then we went to a system that could update any particular page in like sub one minute. Okay.Shawn Wang [00:29:02]: Yeah. Because this is a competitive advantage, right?Jeff Dean [00:29:04]: Because all of a sudden news related queries, you know, if you're, if you've got last month's news index, it's not actually that useful for.Shawn Wang [00:29:11]: News is a special beast. Was there any, like you could have split it onto a separate system.Jeff Dean [00:29:15]: Well, we did. We launched a Google news product, but you also want news related queries that people type into the main index to also be sort of updated.Shawn Wang [00:29:23]: So, yeah, it's interesting. And then you have to like classify whether the page is, you have to decide which pages should be updated and what frequency. Oh yeah.Jeff Dean [00:29:30]: There's a whole like, uh, system behind the scenes that's trying to decide update rates and importance of the pages. So even if the update rate seems low, you might still want to recrawl important pages quite often because, uh, the likelihood they change might be low, but the value of having updated is high.Shawn Wang [00:29:50]: Yeah, yeah, yeah, yeah. Uh, well, you know, yeah. This, uh, you know, mention of latency and, and saving things to this reminds me of one of your classics, which I have to bring up, which is latency numbers. Every programmer should know, uh, was there a, was it just a, just a general story behind that? Did you like just write it down?Jeff Dean [00:30:06]: I mean, this has like sort of eight or 10 different kinds of metrics that are like, how long does a cache mistake? How long does branch mispredict take? How long does a reference domain memory take? How long does it take to send, you know, a packet from the U S to the Netherlands or something? Um,Shawn Wang [00:30:21]: why Netherlands, by the way, or is it, is that because of Chrome?Jeff Dean [00:30:25]: Uh, we had a data center in the Netherlands, um, so, I mean, I think this gets to the point of being able to do the back of the envelope calculations. So these are sort of the raw ingredients of those, and you can use them to say, okay, well, if I need to design a system to do image search and thumb nailing or something of the result page, you know, how, what I do that I could pre-compute the image thumbnails. I could like. Try to thumbnail them on the fly from the larger images. What would that do? How much dis bandwidth than I need? How many des seeks would I do? Um, and you can sort of actually do thought experiments in, you know, 30 seconds or a minute with the sort of, uh, basic, uh, basic numbers at your fingertips. Uh, and then as you sort of build software using higher level libraries, you kind of want to develop the same intuitions for how long does it take to, you know, look up something in this particular kind of.Shawn Wang [00:31:21]: I'll see you next time.Shawn Wang [00:31:51]: Which is a simple byte conversion. That's nothing interesting. I wonder if you have any, if you were to update your...Jeff Dean [00:31:58]: I mean, I think it's really good to think about calculations you're doing in a model, either for training or inference.Jeff Dean [00:32:09]: Often a good way to view that is how much state will you need to bring in from memory, either like on-chip SRAM or HBM from the accelerator. Attached memory or DRAM or over the network. And then how expensive is that data motion relative to the cost of, say, an actual multiply in the matrix multiply unit? And that cost is actually really, really low, right? Because it's order, depending on your precision, I think it's like sub one picodule.Shawn Wang [00:32:50]: Oh, okay. You measure it by energy. Yeah. Yeah.Jeff Dean [00:32:52]: Yeah. I mean, it's all going to be about energy and how do you make the most energy efficient system. And then moving data from the SRAM on the other side of the chip, not even off the off chip, but on the other side of the same chip can be, you know, a thousand picodules. Oh, yeah. And so all of a sudden, this is why your accelerators require batching. Because if you move, like, say, the parameter of a model from SRAM on the, on the chip into the multiplier unit, that's going to cost you a thousand picodules. So you better make use of that, that thing that you moved many, many times with. So that's where the batch dimension comes in. Because all of a sudden, you know, if you have a batch of 256 or something, that's not so bad. But if you have a batch of one, that's really not good.Shawn Wang [00:33:40]: Yeah. Yeah. Right.Jeff Dean [00:33:41]: Because then you paid a thousand picodules in order to do your one picodule multiply.Shawn Wang [00:33:46]: I have never heard an energy-based analysis of batching.Jeff Dean [00:33:50]: Yeah. I mean, that's why people batch. Yeah. Ideally, you'd like to use batch size one because the latency would be great.Shawn Wang [00:33:56]: The best latency.Jeff Dean [00:33:56]: But the energy cost and the compute cost inefficiency that you get is quite large. So, yeah.Shawn Wang [00:34:04]: Is there a similar trick like, like, like you did with, you know, putting everything in memory? Like, you know, I think obviously NVIDIA has caused a lot of waves with betting very hard on SRAM with Grok. I wonder if, like, that's something that you already saw with, with the TPUs, right? Like that, that you had to. Uh, to serve at your scale, uh, you probably sort of saw that coming. Like what, what, what hardware, uh, innovations or insights were formed because of what you're seeing there?Jeff Dean [00:34:33]: Yeah. I mean, I think, you know, TPUs have this nice, uh, sort of regular structure of 2D or 3D meshes with a bunch of chips connected. Yeah. And each one of those has HBM attached. Um, I think for serving some kinds of models, uh, you know, you, you pay a lot higher cost. Uh, and time latency, um, bringing things in from HBM than you do bringing them in from, uh, SRAM on the chip. So if you have a small enough model, you can actually do model parallelism, spread it out over lots of chips and you actually get quite good throughput improvements and latency improvements from doing that. And so you're now sort of striping your smallish scale model over say 16 or 64 chips. Uh, but as if you do that and it all fits in. In SRAM, uh, that can be a big win. So yeah, that's not a surprise, but it is a good technique.Alessio Fanelli [00:35:27]: Yeah. What about the TPU design? Like how much do you decide where the improvements have to go? So like, this is like a good example of like, is there a way to bring the thousand picojoules down to 50? Like, is it worth designing a new chip to do that? The extreme is like when people say, oh, you should burn the model on the ASIC and that's kind of like the most extreme thing. How much of it? Is it worth doing an hardware when things change so quickly? Like what was the internal discussion? Yeah.Jeff Dean [00:35:57]: I mean, we, we have a lot of interaction between say the TPU chip design architecture team and the sort of higher level modeling, uh, experts, because you really want to take advantage of being able to co-design what should future TPUs look like based on where we think the sort of ML research puck is going, uh, in some sense, because, uh, you know, as a hardware designer for ML and in particular, you're trying to design a chip starting today and that design might take two years before it even lands in a data center. And then it has to sort of be a reasonable lifetime of the chip to take you three, four or five years. So you're trying to predict two to six years out where, what ML computations will people want to run two to six years out in a very fast changing field. And so having people with interest. Interesting ML research ideas of things we think will start to work in that timeframe or will be more important in that timeframe, uh, really enables us to then get, you know, interesting hardware features put into, you know, TPU N plus two, where TPU N is what we have today.Shawn Wang [00:37:10]: Oh, the cycle time is plus two.Jeff Dean [00:37:12]: Roughly. Wow. Because, uh, I mean, sometimes you can squeeze some changes into N plus one, but, you know, bigger changes are going to require the chip. Yeah. Design be earlier in its lifetime design process. Um, so whenever we can do that, it's generally good. And sometimes you can put in speculative features that maybe won't cost you much chip area, but if it works out, it would make something, you know, 10 times as fast. And if it doesn't work out, well, you burned a little bit of tiny amount of your chip area on that thing, but it's not that big a deal. Uh, sometimes it's a very big change and we want to be pretty sure this is going to work out. So we'll do like lots of carefulness. Uh, ML experimentation to show us, uh, this is actually the, the way we want to go. Yeah.Alessio Fanelli [00:37:58]: Is there a reverse of like, we already committed to this chip design so we can not take the model architecture that way because it doesn't quite fit?Jeff Dean [00:38:06]: Yeah. I mean, you, you definitely have things where you're going to adapt what the model architecture looks like so that they're efficient on the chips that you're going to have for both training and inference of that, of that, uh, generation of model. So I think it kind of goes both ways. Um, you know, sometimes you can take advantage of, you know, lower precision things that are coming in a future generation. So you can, might train it at that lower precision, even if the current generation doesn't quite do that. Mm.Shawn Wang [00:38:40]: Yeah. How low can we go in precision?Jeff Dean [00:38:43]: Because people are saying like ternary is like, uh, yeah, I mean, I'm a big fan of very low precision because I think that gets, that saves you a tremendous amount of time. Right. Because it's picojoules per bit that you're transferring and reducing the number of bits is a really good way to, to reduce that. Um, you know, I think people have gotten a lot of luck, uh, mileage out of having very low bit precision things, but then having scaling factors that apply to a whole bunch of, uh, those, those weights. Scaling. How does it, how does it, okay.Shawn Wang [00:39:15]: Interesting. You, so low, low precision, but scaled up weights. Yeah. Huh. Yeah. Never considered that. Yeah. Interesting. Uh, w w while we're on this topic, you know, I think there's a lot of, um, uh, this, the concept of precision at all is weird when we're sampling, you know, uh, we just, at the end of this, we're going to have all these like chips that I'll do like very good math. And then we're just going to throw a random number generator at the start. So, I mean, there's a movement towards, uh, energy based, uh, models and processors. I'm just curious if you've, obviously you've thought about it, but like, what's your commentary?Jeff Dean [00:39:50]: Yeah. I mean, I think. There's a bunch of interesting trends though. Energy based models is one, you know, diffusion based models, which don't sort of sequentially decode tokens is another, um, you know, speculative decoding is a way that you can get sort of an equivalent, very small.Shawn Wang [00:40:06]: Draft.Jeff Dean [00:40:07]: Batch factor, uh, for like you predict eight tokens out and that enables you to sort of increase the effective batch size of what you're doing by a factor of eight, even, and then you maybe accept five or six of those tokens. So you get. A five, a five X improvement in the amortization of moving weights, uh, into the multipliers to do the prediction for the, the tokens. So these are all really good techniques and I think it's really good to look at them from the lens of, uh, energy, real energy, not energy based models, um, and, and also latency and throughput, right? If you look at things from that lens, that sort of guides you to. Two solutions that are gonna be, uh, you know, better from, uh, you know, being able to serve larger models or, you know, equivalent size models more cheaply and with lower latency.Shawn Wang [00:41:03]: Yeah. Well, I think, I think I, um, it's appealing intellectually, uh, haven't seen it like really hit the mainstream, but, um, I do think that, uh, there's some poetry in the sense that, uh, you know, we don't have to do, uh, a lot of shenanigans if like we fundamentally. Design it into the hardware. Yeah, yeah.Jeff Dean [00:41:23]: I mean, I think there's still a, there's also sort of the more exotic things like analog based, uh, uh, computing substrates as opposed to digital ones. Uh, I'm, you know, I think those are super interesting cause they can be potentially low power. Uh, but I think you often end up wanting to interface that with digital systems and you end up losing a lot of the power advantages in the digital to analog and analog to digital conversions. You end up doing, uh, at the sort of boundaries. And periphery of that system. Um, I still think there's a tremendous distance we can go from where we are today in terms of energy efficiency with sort of, uh, much better and specialized hardware for the models we care about.Shawn Wang [00:42:05]: Yeah.Alessio Fanelli [00:42:06]: Um, any other interesting research ideas that you've seen, or like maybe things that you cannot pursue a Google that you would be interested in seeing researchers take a step at, I guess you have a lot of researchers. Yeah, I guess you have enough, but our, our research.Jeff Dean [00:42:21]: Our research portfolio is pretty broad. I would say, um, I mean, I think, uh, in terms of research directions, there's a whole bunch of, uh, you know, open problems and how do you make these models reliable and able to do much longer, kind of, uh, more complex tasks that have lots of subtasks. How do you orchestrate, you know, maybe one model that's using other models as tools in order to sort of build, uh, things that can accomplish, uh, you know, much more. Yeah. Significant pieces of work, uh, collectively, then you would ask a single model to do. Um, so that's super interesting. How do you get more verifiable, uh, you know, how do you get RL to work for non-verifiable domains? I think it's a pretty interesting open problem because I think that would broaden out the capabilities of the models, the improvements that you're seeing in both math and coding. Uh, if we could apply those to other less verifiable domains, because we've come up with RL techniques that actually enable us to do that. Uh, effectively, that would, that would really make the models improve quite a lot. I think.Alessio Fanelli [00:43:26]: I'm curious, like when we had Noam Brown on the podcast, he said, um, they already proved you can do it with deep research. Um, you kind of have it with AI mode in a way it's not verifiable. I'm curious if there's any thread that you think is interesting there. Like what is it? Both are like information retrieval of JSON. So I wonder if it's like the retrieval is like the verifiable part. That you can score or what are like, yeah, yeah. How, how would you model that, that problem?Jeff Dean [00:43:55]: Yeah. I mean, I think there are ways of having other models that can evaluate the results of what a first model did, maybe even retrieving. Can you have another model that says, is this things, are these things you retrieved relevant? Or can you rate these 2000 things you retrieved to assess which ones are the 50 most relevant or something? Um, I think those kinds of techniques are actually quite effective. Sometimes I can even be the same model, just prompted differently to be a, you know, a critic as opposed to a, uh, actual retrieval system. Yeah.Shawn Wang [00:44:28]: Um, I do think like there, there is that, that weird cliff where like, it feels like we've done the easy stuff and then now it's, but it always feels like that every year. It's like, oh, like we know, we know, and the next part is super hard and nobody's figured it out. And, uh, exactly with this RLVR thing where like everyone's talking about, well, okay, how do we. the next stage of the non-verifiable stuff. And everyone's like, I don't know, you know, Ellen judge.Jeff Dean [00:44:56]: I mean, I feel like the nice thing about this field is there's lots and lots of smart people thinking about creative solutions to some of the problems that we all see. Uh, because I think everyone sort of sees that the models, you know, are great at some things and they fall down around the edges of those things and, and are not as capable as we'd like in those areas. And then coming up with good techniques and trying those. And seeing which ones actually make a difference is sort of what the whole research aspect of this field is, is pushing forward. And I think that's why it's super interesting. You know, if you think about two years ago, we were struggling with GSM, eight K problems, right? Like, you know, Fred has two rabbits. He gets three more rabbits. How many rabbits does he have? That's a pretty far cry from the kinds of mathematics that the models can, and now you're doing IMO and Erdos problems in pure language. Yeah. Yeah. Pure language. So that is a really, really amazing jump in capabilities in, you know, in a year and a half or something. And I think, um, for other areas, it'd be great if we could make that kind of leap. Uh, and you know, we don't exactly see how to do it for some, some areas, but we do see it for some other areas and we're going to work hard on making that better. Yeah.Shawn Wang [00:46:13]: Yeah.Alessio Fanelli [00:46:14]: Like YouTube thumbnail generation. That would be very helpful. We need that. That would be AGI. We need that.Shawn Wang [00:46:20]: That would be. As far as content creators go.Jeff Dean [00:46:22]: I guess I'm not a YouTube creator, so I don't care that much about that problem, but I guess, uh, many people do.Shawn Wang [00:46:27]: It does. Yeah. It doesn't, it doesn't matter. People do judge books by their covers as it turns out. Um, uh, just to draw a bit on the IMO goal. Um, I'm still not over the fact that a year ago we had alpha proof and alpha geometry and all those things. And then this year we were like, screw that we'll just chuck it into Gemini. Yeah. What's your reflection? Like, I think this, this question about. Like the merger of like symbolic systems and like, and, and LMS, uh, was a very much core belief. And then somewhere along the line, people would just said, Nope, we'll just all do it in the LLM.Jeff Dean [00:47:02]: Yeah. I mean, I think it makes a lot of sense to me because, you know, humans manipulate symbols, but we probably don't have like a symbolic representation in our heads. Right. We have some distributed representation that is neural net, like in some way of lots of different neurons. And activation patterns firing when we see certain things and that enables us to reason and plan and, you know, do chains of thought and, you know, roll them back now that, that approach for solving the problem doesn't seem like it's going to work. I'm going to try this one. And, you know, in a lot of ways we're emulating what we intuitively think, uh, is happening inside real brains in neural net based models. So it never made sense to me to have like completely separate. Uh, discrete, uh, symbolic things, and then a completely different way of, of, uh, you know, thinking about those things.Shawn Wang [00:47:59]: Interesting. Yeah. Uh, I mean, it's maybe seems obvious to you, but it wasn't obvious to me a year ago. Yeah.Jeff Dean [00:48:06]: I mean, I do think like that IMO with, you know, translating to lean and using lean and then the next year and also a specialized geometry model. And then this year switching to a single unified model. That is roughly the production model with a little bit more inference budget, uh, is actually, you know, quite good because it shows you that the capabilities of that general model have improved dramatically and, and now you don't need the specialized model. This is actually sort of very similar to the 2013 to 16 era of machine learning, right? Like it used to be, people would train separate models for lots of different, each different problem, right? I have, I want to recognize street signs and something. So I train a street sign. Recognition recognition model, or I want to, you know, decode speech recognition. I have a speech model, right? I think now the era of unified models that do everything is really upon us. And the question is how well do those models generalize to new things they've never been asked to do and they're getting better and better.Shawn Wang [00:49:10]: And you don't need domain experts. Like one of my, uh, so I interviewed ETA who was on, who was on that team. Uh, and he was like, yeah, I, I don't know how they work. I don't know where the IMO competition was held. I don't know the rules of it. I just trained the models, the training models. Yeah. Yeah. And it's kind of interesting that like people with these, this like universal skill set of just like machine learning, you just give them data and give them enough compute and they can kind of tackle any task, which is the bitter lesson, I guess. I don't know. Yeah.Jeff Dean [00:49:39]: I mean, I think, uh, general models, uh, will win out over specialized ones in most cases.Shawn Wang [00:49:45]: Uh, so I want to push there a bit. I think there's one hole here, which is like, uh. There's this concept of like, uh, maybe capacity of a model, like abstractly a model can only contain the number of bits that it has. And, uh, and so it, you know, God knows like Gemini pro is like one to 10 trillion parameters. We don't know, but, uh, the Gemma models, for example, right? Like a lot of people want like the open source local models that are like that, that, that, and, and, uh, they have some knowledge, which is not necessary, right? Like they can't know everything like, like you have the. The luxury of you have the big model and big model should be able to capable of everything. But like when, when you're distilling and you're going down to the small models, you know, you're actually memorizing things that are not useful. Yeah. And so like, how do we, I guess, do we want to extract that? Can we, can we divorce knowledge from reasoning, you know?Jeff Dean [00:50:38]: Yeah. I mean, I think you do want the model to be most effective at reasoning if it can retrieve things, right? Because having the model devote precious parameter space. To remembering obscure facts that could be looked up is actually not the best use of that parameter space, right? Like you might prefer something that is more generally useful in more settings than this obscure fact that it has. Um, so I think that's always attention at the same time. You also don't want your model to be kind of completely detached from, you know, knowing stuff about the world, right? Like it's probably useful to know how long the golden gate be. Bridges just as a general sense of like how long are bridges, right? And, uh, it should have that kind of knowledge. It maybe doesn't need to know how long some teeny little bridge in some other more obscure part of the world is, but, uh, it does help it to have a fair bit of world knowledge and the bigger your model is, the more you can have. Uh, but I do think combining retrieval with sort of reasoning and making the model really good at doing multiple stages of retrieval. Yeah.Shawn Wang [00:51:49]: And reasoning through the intermediate retrieval results is going to be a, a pretty effective way of making the model seem much more capable, because if you think about, say, a personal Gemini, yeah, right?Jeff Dean [00:52:01]: Like we're not going to train Gemini on my email. Probably we'd rather have a single model that, uh, we can then use and use being able to retrieve from my email as a tool and have the model reason about it and retrieve from my photos or whatever, uh, and then make use of that and have multiple. Um, you know, uh, stages of interaction. that makes sense.Alessio Fanelli [00:52:24]: Do you think the vertical models are like, uh, interesting pursuit? Like when people are like, oh, we're building the best healthcare LLM, we're building the best law LLM, are those kind of like short-term stopgaps or?Jeff Dean [00:52:37]: No, I mean, I think, I think vertical models are interesting. Like you want them to start from a pretty good base model, but then you can sort of, uh, sort of viewing them, view them as enriching the data. Data distribution for that particular vertical domain for healthcare, say, um, we're probably not going to train or for say robotics. We're probably not going to train Gemini on all possible robotics data. We, you could train it on because we want it to have a balanced set of capabilities. Um, so we'll expose it to some robotics data, but if you're trying to build a really, really good robotics model, you're going to want to start with that and then train it on more robotics data. And then maybe that would. It's multilingual translation capability, but improve its robotics capabilities. And we're always making these kind of, uh, you know, trade-offs in the data mix that we train the base Gemini models on. You know, we'd love to include data from 200 more languages and as much data as we have for those languages, but that's going to displace some other capabilities of the model. It won't be as good at, um, you know, Pearl programming, you know, it'll still be good at Python programming. Cause we'll include it. Enough. Of that, but there's other long tail computer languages or coding capabilities that it may suffer on or multi, uh, multimodal reasoning capabilities may suffer. Cause we didn't get to expose it to as much data there, but it's really good at multilingual things. So I, I think some combination of specialized models, maybe more modular models. So it'd be nice to have the capability to have those 200 languages, plus this awesome robotics model, plus this awesome healthcare, uh, module that all can be knitted together to work in concert and called upon in different circumstances. Right? Like if I have a health related thing, then it should enable using this health module in conjunction with the main base model to be even better at those kinds of things. Yeah.Shawn Wang [00:54:36]: Installable knowledge. Yeah.Jeff Dean [00:54:37]: Right.Shawn Wang [00:54:38]: Just download as a, as a package.Jeff Dean [00:54:39]: And some of that installable stuff can come from retrieval, but some of it probably should come from preloaded training on, you know, uh, a hundred billion tokens or a trillion tokens of health data. Yeah.Shawn Wang [00:54:51]: And for listeners, I think, uh, I will highlight the Gemma three end paper where they, there was a little bit of that, I think. Yeah.Alessio Fanelli [00:54:56]: Yeah. I guess the question is like, how many billions of tokens do you need to outpace the frontier model improvements? You know, it's like, if I have to make this model better healthcare and the main. Gemini model is still improving. Do I need 50 billion tokens? Can I do it with a hundred, if I need a trillion healthcare tokens, it's like, they're probably not out there that you don't have, you know, I think that's really like the.Jeff Dean [00:55:21]: Well, I mean, I think healthcare is a particularly challenging domain, so there's a lot of healthcare data that, you know, we don't have access to appropriately, but there's a lot of, you know, uh, healthcare organizations that want to train models on their own data. That is not public healthcare data, uh, not public health. But public healthcare data. Um, so I think there are opportunities there to say, partner with a large healthcare organization and train models for their use that are going to be, you know, more bespoke, but probably, uh, might be better than a general model trained on say, public data. Yeah.Shawn Wang [00:55:58]: Yeah. I, I believe, uh, by the way, also this is like somewhat related to the language conversation. Uh, I think one of your, your favorite examples was you can put a low resource language in the context and it just learns. Yeah.Jeff Dean [00:56:09]: Oh, yeah, I think the example we used was Calamon, which is truly low resource because it's only spoken by, I think 120 people in the world and there's no written text.Shawn Wang [00:56:20]: So, yeah. So you can just do it that way. Just put it in the context. Yeah. Yeah. But I think your whole data set in the context, right.Jeff Dean [00:56:27]: If you, if you take a language like, uh, you know, Somali or something, there is a fair bit of Somali text in the world that, uh, or Ethiopian Amharic or something, um, you know, we probably. Yeah. Are not putting all the data from those languages into the Gemini based training. We put some of it, but if you put more of it, you'll improve the capabilities of those models.Shawn Wang [00:56:49]: Yeah.Jeff Dean [00:56:49]:

Guest: Dr. Sharon Lewin is the Director of the Peter Doherty Institute for Infection and Immunity, where her team studies HIV. She talks about the current landscape in HIV research and treatments, and how new therapies could target latent viral reservoirs. Featured Products and Resources: Register now for IMMUNOLOGY2026! Make the Easy Choice. Try EasySep to Win! The Immunology Science Round Up Immunosurveillance in the Skin: A neuro-epithelial axis can tune regional immunosurveillance against melanoma. B Cells in Aging: B cells contributed to the age-related reduction of naive CD4 T cells. The Gut–Brain Axis in Parkinson’s: Muscularis macrophages, housekeepers of intestinal homeostasis, modulate α-synuclein pathology and neurodegeneration in models of Parkinson’s disease. How IL-2 Signaling Regulates Inflammation: IL-2 signaling promotes the generation of IL-10pos age-associated B cells, with implications for autoimmunity and inflammation. Image courtesy of Dr. Sharon Lewin Subscribe to our newsletter! Never miss updates about new episodes. Subscribe

Topics covered in this episode: Command Book App uvx.sh: Install Python tools without uv or Python Ending 15 years of subprocess polling monty: A minimal, secure Python interpreter written in Rust for use by AI Extras Joke Watch on YouTube About the show Sponsored by us! Support our work through: Our courses at Talk Python Training The Complete pytest Course Patreon Supporters Connect with the hosts Michael: @mkennedy@fosstodon.org / @mkennedy.codes (bsky) Brian: @brianokken@fosstodon.org / @brianokken.bsky.social Show: @pythonbytes@fosstodon.org / @pythonbytes.fm (bsky) Join us on YouTube at pythonbytes.fm/live to be part of the audience. Usually Monday at 10am PT. Older video versions available there too. Finally, if you want an artisanal, hand-crafted digest of every week of the show notes in email form? Add your name and email to our friends of the show list, we'll never share it. Michael #1: Command Book App New app from Michael Command Book App is a native macOS app for developers, data scientists, AI enthusiasts and more. This is a tool I've been using lately to help build Talk Python, Python Bytes, Talk Python Training, and many more applications. It's a bit like advanced terminal commands or complex shell aliases, but hosted outside of your terminal. This leaves the terminal there for interactive commands, exploration, short actions. Command Book manages commands like "tail this log while I'm developing the app", "Run the dev web server with true auto-reload", and even "Run MongoDB in Docker with exactly the settings I need" I'd love it if you gave it a look, shared it with your team, and send me feedback. Has a free version and paid version. Build with Swift and Swift UI Check it out at https://commandbookapp.com Brian #2: uvx.sh: Install Python tools without uv or Python Tim Hopper Michael #3: Ending 15 years of subprocess polling by Giampaolo Rodola The standard library's subprocess module has relied on a busy-loop polling approach since the timeout parameter was added to Popen.wait() in Python 3.3, around 15 years ago The problem with busy-polling CPU wake-ups: even with exponential backoff (starting at 0.1ms, capping at 40ms), the system constantly wakes up to check process status, wasting CPU cycles and draining batteries. Latency: there's always a gap between when a process actually terminates and when you detect it. Scalability: monitoring many processes simultaneously magnifies all of the above. + L1/L2 CPU cache invalidations It's interesting to note that waiting via poll() (or kqueue()) puts the process into the exact same sleeping state as a plain time.sleep() call. From the kernel's perspective, both are interruptible sleeps. Here is the merged PR for this change. Brian #4: monty: A minimal, secure Python interpreter written in Rust for use by AI Samuel Colvin and others at Pydantic Still experimental “Monty avoids the cost, latency, complexity and general faff of using a full container based sandbox for running LLM generated code. “ “Instead, it lets you safely run Python code written by an LLM embedded in your agent, with startup times measured in single digit microseconds not hundreds of milliseconds.” Extras Brian: Expertise is the art of ignoring - Kevin Renskers You don't need to master the language. You need to master your slice. Learning everything up front is wasted effort. Experience changes what you pay attention to. I hate fish - Rands (Michael Lopp) Really about productivity systems And a nice process for dealing with email Michael: Talk Python now has a CLI New essay: It's not vibe coding - Agentic engineering GitHub is having a day Python 3.14.3 and 3.13.12 are available Wall Street just lost $285 billion because of 13 markdown files Joke: Silence, current side project!

We welcome back Andrew Sillifant, Solution Director at Pure Storage, for a deep dive into the concept of data gravity. We start with the traditional 2010 definition coined by Dave McCrory—that data accumulates, making it harder to move, and forcing dependent systems to cluster nearby. However, Andrew presents his core thesis, arguing that this foundational principle is no longer sufficient in a world of exploding complexity. Our conversation emphasizes the need to re-examine data gravity through a modern lens, acknowledging the massive shift to cloud computing and the proliferation of interconnected systems over the last decade. Andrew introduces five crucial dimensions that now describe data's impact: Volume, redefined by context and classification; Dependency, now accelerated by API calls, integration points, and AI agents; Criticality, which includes regulations, security, and implicit SLAs; Velocity, measured by how many functions data is used for; and Latency, complicated by geographic requirements that skew response times. These dimensions highlight how non-physical constraints, like egress fees and data sovereignty laws, create artificial friction that compounds the problem beyond sheer data size. Our discussion concludes with a new framework of five sources of data gravity that IT leaders must address: Technical Gravity (the physical component and mobility), Economic Gravity (the costs of hosting and moving data, like egress fees), Regulatory Gravity (compliance and legal restrictions), Institutional Gravity (the dependency on a small number of people who know how to manage old systems), and Measurement Gravity (budgeting and decision-making risks). Finally, Andrew connects these challenges to Pure Storage, noting how platform features like deduplication and continuous innovation are actively working to lessen the effects of data gravity for customers. To learn more, visit https://blog.purestorage.com/purely-technical/the-economics-of-data-gravity/ Check out the new Pure Storage digital customer community to join the conversation with peers and Pure experts: https://purecommunity.purestorage.com/ 00:00 Intro and Welcome 01:05 Andrew Observations About the USA 04:19 Defining Data Gravity 07:30 Challenges Caused By Data Gravity 09:01 Real World Data Gravity Examples 17:15 Data Gravity Impact Vectors 33:02 New Dimensions of Data Gravity 40:30 Where Pure Helps with Data Gravity

Technology sits at the heart of modern capital markets, and nowhere is that more evident than at the JSE. In the latest episode of Meet the CIO, TechCentral editor Duncan McLeod sits down with Tebalo Tsoaeli, the bourse's CIO, to unpack how technology underpins Africa's largest stock exchange – and how it is evolving for a more digital, global and real-time future. Meet the CIO is brought to you by NTT DATA, where global experience meets local impact. Tsoaeli has spent his entire career in financial services technology, starting out as an application developer at Rand Merchant Bank before holding senior technology roles at Standard Bank, Investec, Nedbank, FirstRand and Sanlam. He became CIO of the JSE three years ago, bringing deep experience in large-scale, mission-critical systems to one of the most tightly regulated technology environments in the country. In the conversation, Tsoaeli reflects on his early exposure to computing and how a formal grounding in computer science shaped his career path. While he is clearly a technologist at heart, he explains how his role has evolved beyond pure IT delivery to focus on strategy, resilience, regulatory compliance and enabling market growth. A major theme of the discussion is the JSE's technology stack and how it has changed over time. Tsoaeli explains how the exchange now works closely with Amazon Web Services, moving away from a purely on-premises model to leverage cloud infrastructure for scalability, resilience and performance. He also addresses the question many market participants ask: can the cloud really be trusted with mission-critical exchange workloads, especially in a world where outages at global providers can have far-reaching consequences? Latency and real-time trading are central concerns for any exchange, and Tsoaeli provides insight into how the JSE's infrastructure supports the full trading lifecycle – from pre-market activity through live trading to post-trade clearing and settlement. He also touches on the exchange's networking architecture and how it is designed to deliver predictable, low-latency performance for brokers and market participants. The episode also explores the JSE's strategic technology partnership with Nasdaq. Tsoaeli explains how this relationship operates at a technology level and what it has delivered so far, including support for market modernisation and international interoperability. Closely linked to this is the modernisation of the JSE's Broker Dealer Accounting system, a project Tsoaeli describes as critical to improving efficiency, resilience and future-readiness. Given the highly regulated nature of financial markets, security and compliance are never far from the conversation. Tsoaeli outlines how the JSE balances innovation with stringent regulatory requirements, and what this means for data protection, operational risk and trust in the market. Looking ahead, the discussion touches on cross-border capital flows, dual listings and the potential role of emerging technologies such as artificial intelligence and machine learning in trading and market operations – along with the risks that come with them. Finally, Tsoaeli shares his perspective on what success looks like for the JSE's technology journey over the next three to five years, how he sees the role of the CIO evolving, and – in a lighter moment – his favourite productivity hack. TechCentral

From Pepcom at CES 2026 in Las Vegas, Paul Tan, COO for Keychron, about the new Q Ultra Series keyboards, their best yet. Highlights include an 8K wireless polling rate with ultra-low latency, exceptional battery life measured in months, a premium CNC-milled aluminum build, upgraded switches, and browser-based configuration that eliminates software downloads that allow for deep customization.  Show Notes: Chapters: Links: Guests: Support:      Become a MacVoices Patron on Patreon     http://patreon.com/macvoices      Enjoy this episode? Make a one-time donation with PayPal Connect:      Web:     http://macvoices.com      Twitter:     http://www.twitter.com/chuckjoiner     http://www.twitter.com/macvoices      Mastodon:     https://mastodon.cloud/@chuckjoiner      Facebook:     http://www.facebook.com/chuck.joiner      MacVoices Page on Facebook:     http://www.facebook.com/macvoices/      MacVoices Group on Facebook:     http://www.facebook.com/groups/macvoice      LinkedIn:     https://www.linkedin.com/in/chuckjoiner/      Instagram:     https://www.instagram.com/chuckjoiner/ Subscribe:      Audio in iTunes     Video in iTunes      Subscribe manually via iTunes or any podcatcher:      Audio: http://www.macvoices.com/rss/macvoicesrss      Video: http://www.macvoices.com/rss/macvoicesvideorss

From Pepcom at CES 2026 in Las Vegas, Paul Tan, COO for Keychron, about the new Q Ultra Series keyboards, their best yet. Highlights include an 8K wireless polling rate with ultra-low latency, exceptional battery life measured in months, a premium CNC-milled aluminum build, upgraded switches, and browser-based configuration that eliminates software downloads that allow for deep customization.  Show Notes: Chapters: Links: Guests: Support:      Become a MacVoices Patron on Patreon      http://patreon.com/macvoices      Enjoy this episode? Make a one-time donation with PayPal Connect:      Web:      http://macvoices.com      Twitter:      http://www.twitter.com/chuckjoiner      http://www.twitter.com/macvoices      Mastodon:      https://mastodon.cloud/@chuckjoiner      Facebook:      http://www.facebook.com/chuck.joiner      MacVoices Page on Facebook:      http://www.facebook.com/macvoices/      MacVoices Group on Facebook:      http://www.facebook.com/groups/macvoice      LinkedIn:      https://www.linkedin.com/in/chuckjoiner/      Instagram:      https://www.instagram.com/chuckjoiner/ Subscribe:      Audio in iTunes      Video in iTunes      Subscribe manually via iTunes or any podcatcher:      Audio: http://www.macvoices.com/rss/macvoicesrss      Video: http://www.macvoices.com/rss/macvoicesvideorss

Snow in the forecast, bass in the bloodstream. We kick things off with a little storm prep and then slide straight into three new releases that light up different corners of the pop universe: a girl band debut with real instruments and real restraint, a solo star leaning into neon funk, and a multinational rookie squad ringing the alarm with swagger.First, LATENCY's “It Was Love” trades flash for feel. Five members play and sing, shaping a soft, mid-tempo band sound that grows from quiet reflection to gentle resolve. Clean guitars, steady percussion, and a tasteful piano-synth color create space for melodies that stick without shouting. By the time the guitar solo lifts around the midpoint, the song has earned its momentum—intimate, bittersweet, and clear-eyed.Then Kento Nakajima flips the room with “XTC”. Think syncopated 80s synths, jazzy guitar licks, and a rhythm that snaps you into motion. Kenty's vocal control is the secret weapon, gliding from sultry lows to sharp falsetto as the track dances between playful invitation and seductive tension. It's a confident solo statement that blends disco shimmer with modern polish, perfect for late-night drives or living room dance breaks.Finally, meet ALPHA DRIVE ONE and their adrenaline-spiked “FREAK ALARM.” The intro hums like a clock before exploding into 80s hip-hop charge—reverb bass, crisp drums, and a chorus that stretches like a siren. The lyrics double as a manifesto: they're the “new alien,” inviting you to own your edge and step into the circle. It's a debut with grit and charm, made to wake up your playlist.If you want a soundtrack for the week—reflection, heat, and high-voltage confidence—this one's for you. Listen, save your favorite moments, and tell us which chorus won your day.LATENCY Instagram X YouTube It Was LoveKento Nakajima Instagram X YouTube XTCALPHA DRIVE ONE Instagram X YouTube FREAK ALARMSupport the showPlease help Music Elixir by rating, reviewing, and sharing the episode. We appreciate your support!Follow us on:TwitterInstagram BlueskyIf have questions, comments, or requests click on our form:Music Elixir FormDJ Panic Blog:OK ASIA

Hey! Alex here, with another weekly AI update! It seems like ThursdAI is taking a new direction, as this is our 3rd show this year, and a 3rd deep dive into topics (previously Ralph, Agent Skills), please let me know if the comments if you like this format. This week's deep dive is into Clawdbot, a personal AI assistant you install on your computer, but can control through your phone, has access to your files, is able to write code, help organize your life, but most importantly, it can self improve. Seeing Wolfred (my Clawdbot) learn to transcribe incoming voice messages blew my mind, and I wanted to share this one with you at length! We had Dan Peguine on the show for the deep dive + both Wolfram and Yam are avid users! This one is not to be missed. If ThursdAI is usually too technical for you, use Claude, and install Clawdbot after you read/listen to the deep dive!Also this week, we read Claude's Constitution that Anthropic released, heard a bunch of new TTS models (some are open source and very impressive) and talked about the new lightspeed coding model GLM 4.7 Flash. First the news, then deep dive, lets go

Cursive Cursing. A Cat Named Beef. Getting Twitchy. Let's Test the Show's Latency. Sour creme on your chalupa is not a euphemism. Emails dash jugs of pee. A Baja Better Time. You Need A Grommet. Imma seal up my hole. The hoity toity mall. Shrodenger's seating. Panky and the brian. But I would have liked to watch you struggle a little bit. Flower Child Something. Belter Passwords with Tom and more on this episode of The Morning Stream. Hosted on Acast. See acast.com/privacy for more information.

Cursive Cursing. A Cat Named Beef. Getting Twitchy. Let's Test the Show's Latency. Sour creme on your chalupa is not a euphemism. Emails dash jugs of pee. A Baja Better Time. You Need A Grommet. Imma seal up my hole. The hoity toity mall. Shrodenger's seating. Panky and the brian. But I would have liked to watch you struggle a little bit. Flower Child Something. Belter Passwords with Tom and more on this episode of The Morning Stream. Hosted on Acast. See acast.com/privacy for more information.

As constraints on energy, water, and permitting collide with exploding demand for AI and compute, a once-fringe idea is moving rapidly toward the center of the conversation: putting data centers in space. Starcloud believes orbital infrastructure isn't science fiction—it's a necessary extension of the global compute stack if scaling is going to continue at anything close to its current pace.Founded by Philip Johnston, Starcloud is building space-based compute systems designed to compete on cost, performance, and scale with terrestrial data centers. The company has already flown a data center–grade GPU in orbit and is now working toward larger, commercially viable systems that could reshape where and how AI is powered. We discuss:How energy and permitting constraints are reshaping the future of computeWhy space-based data centers may be economically inevitable, not optionalWhat Starcloud proved by running an H100 GPU in orbitHow launch costs, watts-per-kilogram, and chip longevity define the real economicsThe national security implications of who controls future compute capacity • Chapters •00:00 - Intro00:50 - The issue with data centers02:20 - Explosion of the data center debates04:58 - Philip's 5GW data center rendering and early conceptions of data centers in space at YC08:16 - Proving people wrong11:17 - The team at Starcloud today12:29 - Competing against SpaceX's data center14:42 - Sam Altman's beef with Starlink16:52 - Economics of Orbital vs Terrestrial Data Centers by Andrew McCallip21:33 - Where are we putting these things?23:50 - Latency in space25:59 - Political side of building data centers28:36 - Starcloud 130:16 - Space based processors30:51 - Shakespeare in space32:00 - Hardening an Nvidia H100 against radiation and making chips in space economical34:43 - Cooling systems in space36:01 - How Starcloud is thinking about replacing failed GPUs38:46 - The mission for Starcloud 240:05 - Competitors outside of SpaceX40:49 - Getting to economical launch costs44:35 - Will the next great wars be over water and power for data centers?46:25 - What keeps Philip up at night?47:11 - What keeps Mo up at night? • Show notes •Starcloud's website — https://www.starcloud.com/Philip's socials — https://x.com/PhilipJohnstonMo's socials — https://x.com/itsmoislamPayload's socials — https://twitter.com/payloadspace / https://www.linkedin.com/company/payloadspaceIgnition's socials — https://twitter.com/ignitionnuclear /  https://www.linkedin.com/company/ignition-nuclear/Tectonic's socials — https://twitter.com/tectonicdefense / https://www.linkedin.com/company/tectonicdefense/Valley of Depth archive — Listen: https://pod.payloadspace.com/ • About us •Valley of Depth is a podcast about the technologies that matter — and the people building them. Brought to you by Arkaea Media, the team behind Payload (space), Ignition (nuclear energy), and Tectonic (defense tech), this show goes beyond headlines and hype. We talk to founders, investors, government officials, and military leaders shaping the future of national security and deep tech. From breakthrough science to strategic policy, we dive into the high-stakes decisions behind the world's hardest technologies.Payload: www.payloadspace.comTectonic: www.tectonicdefense.comIgnition: www.ignition-news.com

"How do I create the 'wow' moment for my end user? And slowness never creates the wow moment." PolarGrid founder and CEO Rade Kovacevic believes GenAI video and voice will be killer apps once they can function in real-time. Enabling real-time GenAI requires uncorking the inference bottleneck that the hyperscalers have helped build. The BetaKit podcast is presented by Fasken Emerging Tech, supporting trailblazing startups, venture capital funds and acquirers of high-growth tech companies for over 30 years. If you're curious about the health of Canada's tech M&A scene, you've got to check out Exit InSights. It's a first-of-its-kind report from Fasken's Emerging Technology & Venture Capital Group that analyses private M&A activity among VC-backed and high-growth tech companies. You'll learn how buyers and sellers are maximizing value, minimizing risk, and navigating one of the most vibrant tech ecosystems out there. Download your free copy of the report.

In this episode, Sahaj Garg, CTO of wispr.ai, joins SE Radio host Robert Blumen to talk about the challenges of building low-latency AI applications. They discuss latency's effect on consumer behavior as well as interactive applications. The conversation explores how to measure latency and how scale impacts it. Then Sahaj and Robert shift to themes around AI, including whether "AI" means LLMs or something broader, as they look at latency requirements and challenges around subtypes of AI applications. The final part of the episode explores techniques for managing latency in AI: speed vs accuracy trade-offs; speed vs cost; latency vs cost; choosing the right model; reducing quantization; distillation; and guessing + validating.

AI was supposed to help humans think better, decide better, and operate with more agency. Instead, many of us feel slower, less confident, and strangely replaceable.In this episode of Design of AI, we interviewed Ovetta Sampson about what quietly went wrong. Not in theory—in practice. We examine how frictionless tools displaced intention, how “freedom” became confused with unlimited capability, and how responsibility dissolved behind abstraction layers, vendors, and models no one fully controls.This is not an anti-AI conversation. It's a reckoning with what happens when adoption outruns judgment.Ovetta Sampson is a tech industry leader who has spent more than a decade leading engineers, designers, and researchers across some of the most influential organizations in technology, including Google, Microsoft, IDEO, and Capital One. She has designed and delivered machine learning, artificial intelligence, and enterprise software systems across multiple industries, and in 2023 was named one of Business Insider's Top 15 People in Enterprise Artificial Intelligence.Join her mailing list⁠ | Right AI | Free Mindful AI Playbook Why 2026 Will Force Teams to Rethink How Much AI They Actually NeedThe risks are no longer abstract. The tradeoffs are no longer subtle. Teams are already feeling the consequences: bloated tool stacks, degraded judgment, unclear accountability, and productivity that looks impressive but feels empty.The next advantage will not come from adding more AI. It will come from removing it deliberately.Organizations that adapt will narrow where AI is used—essential systems, bounded experiments, and clearly protected human decision points. The payoff won't just be cost savings. It will be the return of clarity, ownership, and trust. This is going to manifest first with individuals and small startups who were early adopters of AI. My prediction is that this year they'll start cutting the number of AI models they pay for because the era of experimentation is over and we're now entering a period where deliberate choices will matter more than how fast the model is. Read the full article on LinkedIn. Do You Really Need Frontier Models for Your Product to Work?For most teams, the honest answer is no.Open-source and on-device models already cover the majority of real business needs: internal tooling, retrieval, summarization, classification, workflow automation, and privacy-sensitive systems. The capability gap is routinely overstated—often by those selling access.What open models offer instead is control: over data, cost, latency, deployment, and failure modes. They make accountability visible again. This video explains why the “frontier advantage” is mostly narrative:Independent evaluations now show that open-source AI models can handle most everyday business tasks—summarizing documents, answering questions, drafting content, and internal analysis—at levels comparable to paid systems. The LMSYS Chatbot Arena, which runs blind human comparisons between models, consistently ranks open models close to top proprietary ones.Major consultancies now document why enterprises are switching: predictable costs, data control, and fewer legal and governance risks. McKinsey notes that open models reduce vendor lock-in and compliance exposure in regulated environments.Thanks for reading Design of AI: Strategies for Product Teams & Agencies! Subscribe for free to receive new posts and support my work.What Happens When “Freedom” Becomes an Excuse Not to Set Boundaries?We've confused freedom with capability. If a system can do something, we assume it should. That logic dissolves moral boundaries and replaces responsibility with abstraction: the model did it, the system allowed it.When no one owns the boundary, harm becomes an emergent property instead of a design failure.What If AI Doesn't Have to Be Owned by Corporations?We're going to experience a rise in AI experts challenging the expectations that Silicon Valley should control AI.What if AI doesn't need to be centralized, rented, or governed exclusively by corporate interests?On-device models and open ecosystems offer a different future—less extraction, fewer opaque incentives, and more meaningful choice.Follow Antoine Valot as him and Postcapitalist Design Club explore new ways of liberating AI.Are We Using AI for Anything That Actually Matters?Much of today's AI usage is performative productivity and ego padding that signals relevance while eroding self-trust. We're outsourcing thinking we are still capable of doing ourselves.AI should amplify judgment and creativity. Use this insanely powerful technology to make you achieve greater outcomes, not deliver a higher amount of subpar work to the world.If We Know the Risks Now, Why Are We Still Acting Surprised?The paper “The AI Model Risk Catalog” removes the last excuse.Failure modes are documented. Harms are mapped. Blind spots are known.Continuing to deploy without contingency planning is no longer innovation—it's negligence. If a team can't explain how its system fails safely, who intervenes, and what happens next, it isn't ready for real-world use.If Guardrails Don't Work, What Actually Protects Us?Every AI model and product is at risk of a major attack and exploit.AI systems are structurally vulnerable. The reason we haven't seen a catastrophic failure yet isn't safety—it's limited adoption and permissions.Guardrails fail under pressure. Policies collapse at scale. The only real protection is limiting blast radius: constraining autonomy and refusing to grant authority systems can't safely hold.Why Should Teams Decide Before They Build?The Decision-Forcing AI Business Case Canvas from Unhyped is essential for planning how to leverage AI in your products.Before discussing capabilities, teams must answer:* Who is accountable when this fails?* What judgment must remain human?* What harms are unacceptable—even if the system works?This canvas offers alignment on vision, responsibility, and impact isn't bureaucracy.It's baseline design discipline.Consider the TradeoffsThe conversation with Ovetta Sampson challenges a belief that shaped the last phase of AI adoption: that faster is always better, and that dependence on OpenAI, Google, or Anthropic is inevitable.That belief works during experimentation.It breaks the moment your product starts to matter.As teams scale, speed stops being the constraint. Trust, cost predictability, and accountability take its place. The question shifts from How fast can we ship? to What are we tying our business to—and what happens when it fails?One path optimizes for immediate momentum and simplicity. The other requires more upfront effort, but fundamentally changes where risk, data, and control live.This isn't a technical choice. It's a business one.As usage grows, externalized risk stops being abstract and starts showing up in margins, contracts, and customer trust.As that pressure builds, the impact becomes visible in the product experience itself.Latency creeps in. Costs compound quietly. Outputs vary in ways teams struggle to explain. What once felt powerful starts to feel fragile. Teams spend more time managing side effects than delivering value.At that point, you realize you didn't just choose a model.You chose a UX trajectory.Frontier models feel impressive early, but often lead to expensive, inconsistent experiences over time. Smaller, tuned models trade spectacle for reliability—and reliability is what users actually trust.Eventually, the conversation moves from UX to business fundamentals.Token pricing that felt negligible becomes material. Vendor updates change behavior you didn't choose. Security and compliance questions become harder to abstract away. You realize that outsourcing intelligence also outsourced leverage.This final image makes the tradeoff explicit. Paid frontier models buy speed and simplicity. Open or self-managed approaches buy independence, cost control, and long-term defensibility. Pretending these lead to the same outcomes is the mistake.This transition, from novelty to ownership, is exactly where Right AI Now is focused. Through her consultancy, Ovetta helps teams redesign AI decisions around outcomes that actually matter at scale: customer trust, data sovereignty, operational stability, and long-term value creation.These are also the themes we hear most consistently from the Design of AI audience. Founders and product leaders aren't asking for more tools—they're asking for clearer decisions. They want to know why AI products succeed and fail. We'll be going deeper on this shift throughout 2026, including a rebrand of the podcast, name and all.Improve Your AI ProductIf your organization is at the inflection point where AI needs to deliver real value without eroding trust, this is where I can help you. I've worked with teams at Microsoft, Spotify, and Mozilla to help leaders decide what to build, how to deliver value, and prioritize roadmaps. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit designofai.substack.com

Healthcare organizations are navigating modernization under intense regulatory, security, and resource constraints. This episode explores how the Microsoft technology stack shows up differently in healthcare. The conversation breaks down hybrid cloud realities, Azure managed services, security and compliance, business resiliency, disaster recovery, and cost optimization, all grounded in real healthcare use cases. The episode also explores at how organizations can measure ROI beyond cost savings, connecting Microsoft investments to patient care, clinician experience, and operational resilience. Speakers: Jennifer Johnson, Director of Healthcare at Connection David Carey and Kevin Paiva, Senior Field Solution Architects at Connection Show Notes: 00:10 Welcome and session overview 01:40 Why healthcare cloud adoption is different 02:10 Defining hybrid cloud in healthcare 03:00 Why hybrid is now the default model 03:55 Latency myths and performance realities 04:45 Which workloads belong on-prem vs. in the cloud 05:45 SaaS, staffing pressure, and infrastructure complexity 06:30 Azure managed services and Connection's approach 07:45 Co-managed Azure vs. fully outsourced models 08:30 Why Azure over other hyperscalers 09:20 Azure security, HIPAA, and Zero Trust 10:30 Azure Health Data Services 11:45 Business continuity vs. business resiliency 14:10 What healthcare leaders worry about most today 15:00 Disaster recovery and Azure Expert MSP 16:30 Post-pandemic resource constraints 17:30 Application sprawl, security, and identity management 18:50 Cost containment and ROI in healthcare IT 21:15 The teams behind Connection's Microsoft practice 24:45 Final takeaways and next steps

In this episode of the Crazy Wisdom podcast, host Stewart Alsop interviews Marcin Dymczyk, CPO and co-founder of SevenSense Robotics, exploring the fascinating world of advanced robotics and AI. Their conversation covers the evolution from traditional "standard" robotics with predetermined pathways to advanced robotics that incorporates perception, reasoning, and adaptability - essentially the AGI of physical robotics. Dymczyk explains how his company builds "the eyes and brains of mobile robots" using camera-based autonomy algorithms, drawing parallels between robot sensing systems and human vision, inner ear balance, and proprioception. The discussion ranges from the technical challenges of sensor fusion and world models to broader topics including robotics regulation across different countries, the role of federalism in innovation, and how recent geopolitical changes are driving localized high-tech development, particularly in defense applications. They also touch on the democratization of robotics for small businesses and the philosophical implications of increasingly sophisticated AI systems operating in physical environments. To learn more about SevenSense, visit www.sevensense.ai.Check out this GPT we trained on the conversationTimestamps00:00 Introduction to Robotics and Personal Journey05:27 The Evolution of Robotics: From Standard to Advanced09:56 The Future of Robotics: AI and Automation12:09 The Role of Edge Computing in Robotics17:40 FPGA and AI: The Future of Robotics Processing21:54 Sensing the World: How Robots Perceive Their Environment29:01 Learning from the Physical World: Insights from Robotics33:21 The Intersection of Robotics and Manufacturing35:01 Journey into Robotics: Education and Passion36:41 Practical Robotics Projects for Beginners39:06 Understanding Particle Filters in Robotics40:37 World Models: The Future of AI and Robotics41:51 The Black Box Dilemma in AI and Robotics44:27 Safety and Interpretability in Autonomous Systems49:16 Regulatory Challenges in Robotics and AI51:19 Global Perspectives on Robotics Regulation54:43 The Future of Robotics in Emerging Markets57:38 The Role of Engineers in Modern WarfareKey Insights1. Advanced robotics transcends traditional programming through perception and intelligence. Dymczyk distinguishes between standard robotics that follows rigid, predefined pathways and advanced robotics that incorporates perception and reasoning. This evolution enables robots to make autonomous decisions about navigation and task execution, similar to how humans adapt to unexpected situations rather than following predetermined scripts.2. Camera-based sensing systems mirror human biological navigation. SevenSense Robotics builds "eyes and brains" for mobile robots using multiple cameras (up to eight), IMUs (accelerometers/gyroscopes), and wheel encoders that parallel human vision, inner ear balance, and proprioception. This redundant sensing approach allows robots to navigate even when one system fails, such as operating in dark environments where visual sensors are compromised.3. Edge computing dominates industrial robotics due to connectivity and security constraints. Many industrial applications operate in environments with poor connectivity (like underground grocery stores) or require on-premise solutions for confidentiality. This necessitates powerful local processing capabilities rather than cloud-dependent AI, particularly in automotive factories where data security about new models is paramount.4. Safety regulations create mandatory "kill switches" that bypass AI decision-making. European and US regulatory bodies require deterministic safety systems that can instantly stop robots regardless of AI reasoning. These systems operate like human reflexes, providing immediate responses to obstacles while the main AI brain handles complex navigation and planning tasks.5. Modern robotics development benefits from increasingly affordable optical sensors. The democratization of 3D cameras, laser range finders, and miniature range measurement chips (costing just a few dollars from distributors like DigiKey) enables rapid prototyping and innovation that was previously limited to well-funded research institutions.6. Geopolitical shifts are driving localized high-tech development, particularly in defense applications. The changing role of US global leadership and lessons from Ukraine's drone warfare are motivating countries like Poland to develop indigenous robotics capabilities. Small engineering teams can now create battlefield-effective technology using consumer drones equipped with advanced sensors.7. The future of robotics lies in natural language programming for non-experts. Dymczyk envisions a transformation where small business owners can instruct robots using conversational language rather than complex programming, similar to how AI coding assistants now enable non-programmers to build applications through natural language prompts.

بزاف ديال المهندسين (Engineers) ف الـ Backend كايغلطو ف هاد المفاهيم الأساسية. واش السيستيم ديالك غيبقى خدام إلا تّشق فيه شي "Fault"؟ واش هاديك الـ Moyenne (الأرقام المتوسطة) اللي كاتشوف ف الـ Dashboard هي اللي كاتشرح الحقيقة؟ف هاد الحلقة من سحابة (S7aba)، غادي نغوصو ف الكتاب اللي دار الروينة ف هاد الدومين: Designing Data-Intensive Applications ديال Martin Kleppmann. غادي نجاوبو على هاد الأسئلة

De Rosalia à Little Simz, en passant par Bad Bunny, Sophian Fanen balaye son année 2025. En cette fin d'année, Sophian vous fait plaisir et sélectionne ses obsessions favorites. Rosalía, Berghain, tiré de l'album Lux (Columbia Records, 2025) Bad Bunny, Weltita (feat. Chuwi), tiré de l'album Debí Tirar Más Fotos (Rimas Entertainment)  Theodora, Ils me rient tous au nez, tiré de l'album Mega BBL (Boss Lady, 2025)  Tarta Relena, Si veriash a la rana, tiré de l'album És pregunta (Latency, 2024) Ale hop & Titi Bakorta, Bonne année, tiré de l'album Mapambazuko (Nyege Nyege Tapes, 2025)  Andrea Laszlo de Simone, Quando, tiré de l'album Una Lunghissima Ombra (Ekler/Hamburger Records, 2025)  Blaiz Fayah, Maureen et DJ Glad, Money Pull Up, tiré de l'album Shatta Ting (Creepy Music, 2025)  Little Simz, Lion (feat. Obongjayar), tiré de l'album Lotus (Awal, 2025)  Miki, Jtm encore, tiré de l'album Industry Plant (Structure, 2025)  Xania Monet, How Was I Supposed to Know?, tiré de l'album Unfolded (TMJ/Hallwood, 2025) 

De Rosalia à Little Simz, en passant par Bad Bunny, Sophian Fanen balaye son année 2025. En cette fin d'année, Sophian vous fait plaisir et sélectionne ses obsessions favorites. Rosalía, Berghain, tiré de l'album Lux (Columbia Records, 2025) Bad Bunny, Weltita (feat. Chuwi), tiré de l'album Debí Tirar Más Fotos (Rimas Entertainment)  Theodora, Ils me rient tous au nez, tiré de l'album Mega BBL (Boss Lady, 2025)  Tarta Relena, Si veriash a la rana, tiré de l'album És pregunta (Latency, 2024) Ale hop & Titi Bakorta, Bonne année, tiré de l'album Mapambazuko (Nyege Nyege Tapes, 2025)  Andrea Laszlo de Simone, Quando, tiré de l'album Una Lunghissima Ombra (Ekler/Hamburger Records, 2025)  Blaiz Fayah, Maureen et DJ Glad, Money Pull Up, tiré de l'album Shatta Ting (Creepy Music, 2025)  Little Simz, Lion (feat. Obongjayar), tiré de l'album Lotus (Awal, 2025)  Miki, Jtm encore, tiré de l'album Industry Plant (Structure, 2025)  Xania Monet, How Was I Supposed to Know?, tiré de l'album Unfolded (TMJ/Hallwood, 2025) 

In this episode of The Effortless Podcast, Amit Prakash and Dheeraj Pandey dive deep into one of the most important shifts happening in AI today: the convergence of structured and unstructured data, interfaces, and systems.Together, they unpack how conversations—not CRM fields—hold the real ground truth; why schemas still matter in an AI-driven world; and how agents can evolve into true managers, coaches, and chiefs of staff for revenue teams. They explore the cognitive science behind visual vs conversational UI, the future of dynamically generated interfaces, and the product depth required to build enduring AI-native software.Amit and Dheeraj break down the tension between deterministic and probabilistic systems, the limits of prompt-driven workflows, and why the future of enterprise AI is “both-and” rather than “either-or.” It's a masterclass in modern product, data design, and the psychology of building intelligent tools.Key Topics & Timestamps 00:00 – Introduction02:00 – Why conversations—not CRM fields—hold real ground truth05:00 – Reps as labelers and the parallels with AI training pipelines08:00 – Business logic vs world models: defining meaning inside enterprises11:00 – Prompts flatten nuance; schemas restore structure14:00 – SQL schemas as the true model of a business17:00 – CRM overload and the friction of rigid data entry20:00 – AI agents that debrief and infer fields dynamically23:00 – Capturing qualitative signals: champions, pain, intent26:00 – Multi-source context: transcripts, email threads, Slack29:00 – Why structure is required for math, aggregation, forecasting32:00 – Aggregating unstructured data to reveal organizational issues35:00 – Labels, classification, and the limits of LLM-only workflows38:00 – Deterministic (SQL/Python) vs probabilistic (LLMs) systems41:00 – Transitional workflows: humans + AI field entry44:00 – Trust issues and the confusion of the early AI market47:00 – Avoiding “Clippy moments” in agent design50:00 – Latency, voice UX, and expectations for responsiveness53:00 – Human-machine interface for SDRs vs senior reps56:00 – Structured vs unstructured UI: cognitive science insights59:00 – Charts vs paragraphs: parallel vs sequential processing1:02:00 – The “Indian thali” dashboard problem and dynamic UI1:05:00 – Exploration modes, drill-downs, and empty prompts1:08:00 – Dynamic leaves, static trunk: designing hierarchy1:11:00 – Both-and thinking: voice + visual, structured + unstructured1:14:00 – Why “good enough” AI fails without deep product1:17:00 – PLG, SLG, data access, and trust barriers1:20:00 – Closing reflections and the future of AI-native softwareHosts: Amit Prakash – CEO and Founder at AmpUp, former engineer at Google AdSense and Microsoft Bing, with extensive expertise in distributed systems and machine learningDheeraj Pandey – Co-founder and CEO at DevRev, former Co-founder & CEO of Nutanix. A tech visionary with a deep interest in AI, systems, and the future of work.Follow the Hosts:Amit PrakashLinkedIn – Amit Prakash I LinkedInTwitter/X – https://x.com/amitp42Dheeraj PandeyLinkedIn –Dheeraj Pandey | LinkedIn Twitter/X – https://x.com/dheerajShare your thoughts : Have questions, comments, or ideas for future episodes?Email us at EffortlessPodcastHQ@gmail.comDon't forget to Like, Comment, and Subscribe for more conversations at the intersection of AI, technology, and innovation.

Every few years, the world of product management goes through a phase shift. When I started at Microsoft in the early 2000s, we shipped Office in boxes. Product cycles were long, engineering was expensive, and user research moved at the speed of snail mail. Fast forward a decade and the cloud era reset the speed at which we build, measure, and learn. Then mobile reshaped everything we thought we knew about attention, engagement, and distribution.Now we are standing at the edge of another shift. Not a small shift, but a tectonic one. Artificial intelligence is rewriting the rules of product creation, product discovery, product expectations, and product careers.To help make sense of this moment, I hosted a panel of world class product leaders on the Fireside PM podcast:• Rami Abu-Zahra, Amazon product leader across Kindle, Books, and Prime Video• Todd Beaupre, Product Director at YouTube leading Home and Recommendations• Joe Corkery, CEO and cofounder of Jaide Health • Tom Leung (me), Partner at Palo Alto Foundry• Lauren Nagel, VP Product at Mezmo• David Nydegger, Chief Product Officer at OvivaThese are leaders running massive consumer platforms, high stakes health tech, and fast moving developer tools. The conversation was rich, honest, and filled with specific examples. This post summarizes the discussion, adds my own reflections, and offers a practical guide for early and mid career PMs who want to stay relevant in a world where AI is redefining what great product management looks like.Table of Contents* What AI Cannot Do and Why PM Judgment Still Matters* The New AI Literacy: What PMs Must Know by 2026* Why Building AI Products Speeds Up Some Cycles and Slows Down Others* Whether the PM, Eng, UX Trifecta Still Stands* The Biggest Risks AI Introduces Into Product Development* Actionable Advice for Early and Mid Career PMs* My Takeaways and What Really Matters Going Forward* Closing Thoughts and Coaching Practice1. What AI Cannot Do and Why PM Judgment Still MattersWe opened the panel with a foundational question. As AI becomes more capable every quarter, what is left for humans to do. Where do PMs still add irreplaceable value. It is the question every PM secretly wonders.Todd put it simply: “At the end of the day, you have to make some judgment calls. We are not going to turn that over anytime soon.”This theme came up again and again. AI is phenomenal at synthesizing, drafting, exploring, and narrowing. But it does not have conviction. It does not have lived experience. It does not feel user pain. It does not carry responsibility.Joe from Jaide Health captured it perfectly when he said: “AI cannot feel the pain your users have. It can help meet their goals, but it will not get you that deep understanding.”There is still no replacement for sitting with a frustrated healthcare customer who cannot get their clinical data into your system, or a creator on YouTube who feels the algorithm is punishing their art, or a devops engineer staring at an RCA output that feels 20 percent off.Every PM knows this feeling: the moment when all signals point one way, but your gut tells you the data is incomplete or misleading. This is the craft that AI does not have.Why judgment becomes even more important in an AI worldDavid, who runs product at a regulated health company, said something incredibly important: “Knowing what great looks like becomes more essential, not less. The PM's that thrive in AI are the ones with great product sense.”This is counterintuitive for many. But when the operational work becomes automated, the differentiation shifts toward taste, intuition, sequencing, and prioritization.Lauren asked the million dollar question. “How are we going to train junior PMs if AI is doing the legwork. Who teaches them how to think.”This is a profound point. If AI closes the gap between junior and senior PMs in execution tasks, the difference will emerge almost entirely in judgment. Knowing how to probe user problems. Knowing when a feature is good enough. Knowing which tradeoffs matter. Knowing which flaw is fatal and which is cosmetic.AI is incredible at writing a PRD. AI is terrible at knowing whether the PRD is any good.Which means the future PM becomes more strategic, more intuitive, more customer obsessed, and more willing to make thoughtful bets under uncertainty.2. The New AI Literacy: What PMs Must Know by 2026I asked the panel what AI literacy actually means for PMs. Not the hype. Not the buzzwords. The real work.Instead of giving gimmicky answers, the discussion converged on a clear set of skills that PMs must master.Skill 1: Understanding context engineeringDavid laid this out clearly: “Knowing what LMS are good at and what they are not good at, and knowing how to give them the right context, has become a foundational PM skill.”Most PMs think prompt engineering is about clever phrasing. In reality, the future is about context engineering. Feeding models the right data. Choosing the right constraints. Deciding what to ignore. Curating inputs that shape outputs in reliable ways.Context engineering is to AI product development what Figma was to collaborative design. If you cannot do it, you are not going to be effective.Skill 2: Evals, evals, evalsRami said something that resonated with the entire panel: “Last year was all about prompts. This year is all about evals.”He is right.• How do you build a golden dataset.• How do you evaluate accuracy.• How do you detect drift.• How do you measure hallucination rates.• How do you combine UX evals with model evals.• How do you decide what good looks like.• How do you define safe versus unsafe boundaries.AI evaluation is now a core PM responsibility. Not exclusively. But PMs must understand what engineers are testing for, what failure modes exist, and how to design test sets that reflect the real world.Lauren said her PMs write evals side by side with engineering. That is where the world is going.Skill 3: Knowing when to trust AI output and when to override itTodd noted: “It is one thing to get an answer that sounds good. It is another thing to know if it is actually good.”This is the heart of the role. AI can produce strategic recommendations that look polished, structured, and wise. But the real question is whether they are grounded in reality, aligned with your constraints, and consistent with your product vision.A PM without the ability to tell real insight from confident nonsense will be replaced by someone who can.Skill 4: Understanding the physics of model changesThis one surprised many people, but it was a recurring point.Rami noted: “When you upgrade a model, the outputs can be totally different. The evals start failing. The experience shifts.”PMs must understand:• Models get deprecated• Models drift• Model updates can break well tuned prompts• API pricing has real COGS implications• Latency varies• Context windows vary• Some tasks need agents, some need RAG, some need a small finetuned modelThis is product work now. The PM of 2026 must know these constraints as well as a PM of the cloud era understood database limits or API rate limits.Skill 5: How to construct AI powered prototypes in hours, not weeksIt now takes one afternoon to build something meaningful. Zero code required. Prompt, test, refine. Whether you use Replit, Cursor, Vercel, or sandboxed agents, the speed is shocking.But this makes taste and problem selection even more important. The future PM must be able to quickly validate whether a concept is worth building beyond the demo stage.3. Why Building AI Products Speeds Up Some Cycles and Slows Down OthersThis part of the conversation was fascinating because people expected AI to accelerate everything. The panel had a very different view.Fast: Prototyping and concept validationLauren described how her teams can build working versions of an AI powered Root Cause Analysis feature in days, test it with customers, and get directional feedback immediately.“You can think bigger because the cost of trying things is much lower,” she said.For founders, early PMs, and anyone validating hypotheses, this is liberating. You can test ten ideas in a week. That used to take a quarter.Slow: Productionizing AI featuresThe surprising part is that shipping the V1 of an AI feature is slower than most expect.Joe noted: “You can get prototypes instantly. But turning that into a real product that works reliably is still hard.”Why. Because:• You need evals.• You need monitoring.• You need guardrails.• You need safety reviews.• You need deterministic parts of the workflow.• You need to manage COGS.• You need to design fallbacks.• You need to handle unpredictable inputs.• You need to think about hallucination risk.• You need new UI surfaces for non deterministic outputs.Lauren said bluntly: “Vibe coding is fast. Moving that vibe code to production is still a four month process.”This should be printed on a poster in every AI startup office.Very Slow: Iterating on AI powered featuresAnother counterintuitive point. Many teams ship a great V1 but struggle to improve it significantly afterward.David said their nutrition AI feature launched well but: “We struggled really hard to make it better. Each iteration was easy to try but difficult to improve in a meaningful way.”Why is iteration so difficult.Because model improvements may not translate directly into UX improvements. Users need consistency. Drift creates churn. Small changes in context or prompts can cause large changes in behavior.Teams are learning a hard truth: AI powered features do not behave like typical deterministic product flows. They require new iteration muscles that most orgs do not yet have.4. The PM, Eng, UX Trifecta in the AI EraI asked whether the classic PM, Eng, UX triad is still the right model. The audience was expecting disagreement. The panel was surprisingly aligned.The trifecta is not going anywhereRami put it simply: “We still need experts in all three domains to raise the bar.”Joe added: “AI makes it possible for PMs to do more technical work. But it does not replace engineering. Same for design.”AI blurs the edges of the roles, but it does not collapse them. In fact, each role becomes more valuable because the work becomes more abstract.• PMs focus on judgment, sequencing, evaluation, and customer centric problem framing• Engineers focus on agents, systems, architecture, guardrails, latency, and reliability• Designers focus on dynamic UX, non deterministic UX patterns, and new affordances for AI outputsWhat does changeAI makes the PM-Eng relationship more intense. The backbone of AI features is a combination of model orchestration, evaluation, prompting, and context curation. PMs must be tighter than ever with engineering to design these systems.David noted that his teams focus more on individual talents. Some PMs are great at context engineering. Some designers excel at polishing AI generated layouts. Some engineers are brilliant at prompt chaining. AI reveals strengths quickly.The trifecta remains. The skill distribution within it evolves.5. The Biggest Risks AI Introduces Into Product DevelopmentWhen we asked what scares PMs most about AI, the conversation became blunt and honest. Risk 1: Loss of user trustLauren warned: “If people keep shipping low quality AI features, user trust in AI erodes. And then your good AI product suffers from the skepticism.”This is very real. Many early AI features across industries are low quality, gimmicky, or unreliable. Users quickly learn to distrust these experiences.Which means PMs must resist the pressure to ship before the feature is ready.Risk 2: Skill atrophyTodd shared a story that hit home for many PMs. “Junior folks just want to plug in the prompt and take whatever the AI gives them. That is a recipe for having no job later.”PMs who outsource their thinking to AI will lose their judgment. Judgment cannot be regained easily.This is the silent career killer.Risk 3: Safety hazards in sensitive domainsDavid was direct: “If we have one unsafe output, we have to shut the feature off. We cannot afford even small mistakes.”In healthcare, finance, education, and legal industries, the tolerance for error is near zero. AI must be monitored relentlessly. Human in the loop systems are mandatory. The cycles are slower but the stakes are higher.Risk 4: The high bar for AI compared to humansJoe said something I have thought about for years: “AI is held to a much higher standard than human decision making. Humans make mistakes constantly, but we forgive them. AI makes one mistake and it is unacceptable.”This slows adoption in certain industries and creates unrealistic expectations.Risk 5: Model deprecation and instabilityRami described a real problem AI PMs face: “Models get deprecated faster than they get replaced. The next model is not always GA. Outputs change. Prompts break.”This creates product instability that PMs must anticipate and design around.Risk 6: Differentiation becomes hardI shared this perspective because I see so many early stage startups struggle with it.If your whole product is a wrapper around an LLM, competitors will copy you in a week. The real differentiation will not come from using AI. It will come from how deeply you understand the customer, how you integrate AI with proprietary data, and how you create durable workflows.6. Actionable Advice for Early and Mid Career PMsThis was one of my favorite parts of the panel because the advice was humble, practical, and immediately useful.A. Develop deep user empathy. This will become your biggest differentiator.Lauren said it clearly: “Maintain your empathy. Understand the pain your user really has.”AI makes execution cheap. It makes insight valuable.If you can articulate user pain precisely.If you can differentiate surface friction from underlying need.If you can see around corners.If you can prototype solutions and test them in hours.If you can connect dots between what AI can do and what users need.You will thrive.Tactical steps:• Sit in on customer support calls every week.• Watch 10 user sessions for every feature you own.• Talk to customers until patterns emerge.• Ask “why” five times in every conversation.• Maintain a user pain log and update it constantly.B. Become great at context engineeringThis will matter as much as SQL mattered ten years ago.Action steps:• Practice writing prompts with structured context blocks.• Build a library of prompts that work for your product.• Study how adding, removing, or reordering context changes output.• Learn RAG patterns.• Learn when structured data beats embeddings.• Learn when smaller local models outperform big ones.C. Learn eval frameworksThis is non negotiable.You need to know:• Precision vs recall tradeoffs• How to build golden datasets• How to design scenario based evals for UX• How to test for hallucination• How to monitor drift• How to set quality thresholds• How to build dashboards that reflect real world input distributionsYou do not need to write the code.You do need to define the eval strategy.D. Strengthen your product senseYou cannot outsource product taste.Todd said it best: “Imagine asking AI to generate 20 percent growth for you. It will not tell you what great looks like.”To strengthen your product sense:• Review the best products weekly.• Take screenshots of great UX patterns.• Map user flows from apps you admire.• Break products down into primitives.• Ask yourself why a product decision works.• Predict what great would look like before you design it.The PMs who thrive will be the ones who can recognize magic when they see it.E. Stay curiousRami's closing advice was simple and perfect: “Stay curious. Keep learning. It never gets old.”AI changes monthly. The PM who is excited by new ideas will outperform the PM who clings to old patterns.Practical habits:• Read one AI research paper summary each week.• Follow evaluation and model updates from major vendors.• Build at least one small AI prototype a month.• Join AI PM communities.• Teach juniors what you learn. Nothing accelerates mastery faster.F. Embrace velocity and side projectsTodd said that some of his biggest career breakthroughs came from solving problems on the side.This is more true now than ever.If you have an idea, you can build an MVP over a weekend. If it solves a real problem, someone will notice.G. Stay close to engineeringNot because you need to code, but because AI features require tighter PM engineering collaboration.Learn enough to be dangerous:• How embeddings work• How vector stores behave• What latency tradeoffs exist• How agents chain tasks• How model versioning works• How context limits shape UX• Why some prompts blow up API costsIf you can speak this language, you will earn trust and accelerate cycles.H. Understand the business deeplyJoe's advice was timeless: “Know who pays you and how much they pay. Solve real problems and know the business model.”PMs who understand unit economics, COGS, pricing, and funnel dynamics will stand out.7. Tom's Takeaways and What Really Matters Going ForwardI ended the recording by sharing what I personally believe after moderating this discussion and working closely with a variety of AI teams over the past 2 years.Judgment becomes the most valuable PM skillAs AI gets better at analysis, synthesis, and execution, your value shifts to:• Choosing the right problem• Sequencing decisions• Making 55 45 calls• Understanding user pain• Making tradeoffs• Deciding when good is good enough• Defining success• Communicating vision• Influencing the orgAgents can write specs.LLMs can produce strategies.But only humans can choose the right one and commit.Learning speed becomes a competitive advantageI said this on the panel and I believe it more every month.Because of AI, you now have:• Infinite coaches• Infinite mentors• Infinite experts• Infinite documentation• Infinite learning loopsA PM who learns slowly will not survive the next decade. Curiosity, empathy, and velocity will separate great from goodMany panelists said versions of this. The common pattern was:• Understand users deeply• Combine multiple tools creatively• Move quickly• Learn constantlyThe future rewards generalists with taste, speed, and emotional intelligence.Differentiation requires going beyond wrapper appsThis is one of my biggest concerns for early stage founders. If your entire product is a wrapper around a model, you are vulnerable.Durable value will come from:• Proprietary data• Proprietary workflows• Deep domain insight• Organizational trust• Distribution advantage• Safety and reliability• Integration with existing systemsAI is a component, not a moat.8. Closing ThoughtsHosting this panel made me more optimistic about the future of product management. Not because AI will not change the job. It already has. But because the fundamental craft remains alive.Product management has always been about understanding people, making decisions with incomplete information, telling compelling stories, and guiding teams through ambiguity and being right often.AI accelerates the craft. It amplifies the best PMs and exposes the weak ones. It rewards curiosity, empathy, velocity, and judgment.If you want tailored support on your PM career, leadership journey, or executive path, I offer 1 on 1 career, executive, and product coaching at tomleungcoaching.com.OK team. Let's ship greatness. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit firesidepm.substack.com

The Data Center Boom: Five Trends Engineering Firms Need to Know The data center market is experiencing unprecedented growth, driven by artificial intelligence adoption and changing infrastructure demands. For ACEC member firms, this represents both a substantial business opportunity and a chance to shape critical national infrastructure. ACEC's latest Market Intelligence Brief reveals a market poised to reach $62 billion in design and construction spending by 2029, with implications that extend far beyond traditional data center engineering. The launch of ChatGPT in 2022 marked an inflection point. What began as voice assistants has evolved into sophisticated language learning models that consume dramatically more energy. A standard AI query uses about 0.012 kilowatt-hours, while generating a single high-quality image requires 2.0 kWh—roughly 20 times the daily consumption of a standard LED lightbulb. As weekly ChatGPT users surged from 100 million to 700 million between November 2023 and August 2025, the infrastructure implications became impossible to ignore. AI-driven data center power demand, which stood at just 4 gigawatts in 2024, is projected to reach 123 gigawatts by 2035. Even more striking: 70 percent of data center power demand will be driven by AI workloads. This explosive growth requires engineering solutions at unprecedented scale, from power distribution and backup systems to advanced cooling technologies and grid integration strategies. Public perception about data center water consumption often overlooks important nuances in cooling technology. While mechanical cooling systems have historically consumed significant water resources, newer approaches could dramatically reduce water use. Free air cooling, closed-loop systems, and liquid immersion technologies offer low-water use alternatives, with some methods reducing freshwater consumption by 70 percent or more compared to traditional systems. As Thom Jackson, mechanical engineer and partner at Dunham Engineering, notes: "Most data centers utilize closed loop cooling systems requiring no makeup water and minimal maintenance." The "big four" hyperscale operators—Amazon Web Services, Microsoft Azure, Google Cloud Platform, and Meta—have all committed to becoming water-positive by 2030, replenishing more water than they consume. These commitments are driving innovation in cooling system design and creating opportunities for engineering firms with expertise in sustainable mechanical systems. The days of one-size-fits-all data centers are over. Latency requirements, scalability needs, and proximity to end users are accelerating adoption of diverse building types. Edge data centers bring computing closer to users for real-time applications like IoT and 5G. Hyperscale facilities support massive cloud and AI workloads with 100,000-plus servers. Colocation models enable scalable shared environments for enterprises, while modular designs—prefabricated with integrated power and cooling—offer rapid, cost-effective deployment. Each model presents distinct engineering challenges and opportunities, from specialized HVAC systems and high floor-to-ceiling ratios for hyperscale facilities to distributed infrastructure planning for edge networks. Two emerging trends deserve particular attention. First, the Department of Energy has selected four federal sites to host AI data centers paired with clean energy generation, including small modular reactors (SMRs). The Nuclear Regulatory Commission anticipates at least 25 SMR license applications by 2029, signaling strong demand for nuclear co-location expertise. Second, developers are increasingly exploring adaptive reuse of underutilized office spaces, Brownfield sites, and historical buildings. These locations offer existing utility infrastructure that can reduce construction time and costs, making them attractive alternatives despite some design constraints. Recent federal policy changes are streamlining data center deployment. Executive Order 14318 directs agencies to accelerate environmental reviews and permitting, while revisions to New Source Review under the Clean Air Act could allow construction to begin before air permits are issued. ACEC recently formed the Data Center Task Force to advocate for policies that balance speed, affordability, and national security in data center development, complimenting EO 14318. For engineering firms, site selection expertise has become increasingly valuable. Success hinges on sales and use tax exemptions, existing power and fiber connectivity, effective community engagement, and thorough environmental risk assessment. AI-driven planning tools like UrbanFootprint and ESRI ArcGIS are helping developers evaluate site suitability, identifying opportunities for firms. The data center market offers engineering firms a chance to lead in sustainable design, infrastructure innovation, and strategic planning at a moment when digital infrastructure has become as critical as traditional utilities.  

In this week's episode of Hands-On Tech, Lance asks Mikah Sargent about the pros and cons of using powerline ethernet adapters, and Mikah shares his strong thoughts on these devices. Don't forget to send in your questions for Mikah to answer during the show! hot@twit.tv Host: Mikah Sargent Download or subscribe to Hands-On Tech at https://twit.tv/shows/hands-on-tech Want access to the ad-free video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

In this week's episode of Hands-On Tech, Lance asks Mikah Sargent about the pros and cons of using powerline ethernet adapters, and Mikah shares his strong thoughts on these devices. Don't forget to send in your questions for Mikah to answer during the show! hot@twit.tv Host: Mikah Sargent Download or subscribe to Hands-On Tech at https://twit.tv/shows/hands-on-tech Want access to the ad-free video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

In this week's episode of Hands-On Tech, Lance asks Mikah Sargent about the pros and cons of using powerline ethernet adapters, and Mikah shares his strong thoughts on these devices. Don't forget to send in your questions for Mikah to answer during the show! hot@twit.tv Host: Mikah Sargent Download or subscribe to Hands-On Tech at https://twit.tv/shows/hands-on-tech Want access to the ad-free video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

In this week's episode of Hands-On Tech, Lance asks Mikah Sargent about the pros and cons of using powerline ethernet adapters, and Mikah shares his strong thoughts on these devices. Don't forget to send in your questions for Mikah to answer during the show! hot@twit.tv Host: Mikah Sargent Download or subscribe to Hands-On Tech at https://twit.tv/shows/hands-on-tech Want access to the ad-free video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

In this week's episode of Hands-On Tech, Lance asks Mikah Sargent about the pros and cons of using powerline ethernet adapters, and Mikah shares his strong thoughts on these devices. Don't forget to send in your questions for Mikah to answer during the show! hot@twit.tv Host: Mikah Sargent Download or subscribe to Hands-On Tech at https://twit.tv/shows/hands-on-tech Want access to the ad-free video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

In this week's episode of Hands-On Tech, Lance asks Mikah Sargent about the pros and cons of using powerline ethernet adapters, and Mikah shares his strong thoughts on these devices. Don't forget to send in your questions for Mikah to answer during the show! hot@twit.tv Host: Mikah Sargent Download or subscribe to Hands-On Tech at https://twit.tv/shows/hands-on-tech Want access to the ad-free video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

In this week's episode of Hands-On Tech, Lance asks Mikah Sargent about the pros and cons of using powerline ethernet adapters, and Mikah shares his strong thoughts on these devices. Don't forget to send in your questions for Mikah to answer during the show! hot@twit.tv Host: Mikah Sargent Download or subscribe to Hands-On Tech at https://twit.tv/shows/hands-on-tech Want access to the ad-free video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

In this week's episode of Hands-On Tech, Lance asks Mikah Sargent about the pros and cons of using powerline ethernet adapters, and Mikah shares his strong thoughts on these devices. Don't forget to send in your questions for Mikah to answer during the show! hot@twit.tv Host: Mikah Sargent Download or subscribe to Hands-On Tech at https://twit.tv/shows/hands-on-tech Want access to the ad-free video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

In this week's episode of Hands-On Tech, Lance asks Mikah Sargent about the pros and cons of using powerline ethernet adapters, and Mikah shares his strong thoughts on these devices. Don't forget to send in your questions for Mikah to answer during the show! hot@twit.tv Host: Mikah Sargent Download or subscribe to Hands-On Tech at https://twit.tv/shows/hands-on-tech Want access to the ad-free video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

The Crew and some new and old friends, take on the Mass Driver at Morro Rock. They've made it past the barge and into the compound, now to split up, plant the virus and get out without a bang. Hopefully...Jade, Evan and Peter are joined by Dayeanne Hutton and Anais R Morgan (Infinite Sided Dice) live on stage at San Diego Comic Con 2025!Check out our Youtube Channel for more live panels! HereBTS and art posts will come out for all our patrons to peruse next week!More info can be found here: ⁠⁠⁠⁠⁠⁠linktr.ee/NoLatency⁠⁠⁠⁠⁠⁠If you'd like to support us, We now have a Patreon! ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Patreon.com/nolatency⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Even more information and MERCH is on our website! ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠www.nolatencypodcast.com⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Twitter: @nolatencypodInstagram: @nolatencypodLogo & Map Art By Paris ArrowsmithCharacter Art by: Doodlejumps, Saint and Paris ArrowsmithProducing and Editing by Paris ArrowsmithMusic and Sound sfx by Epidemic Sound.Find @SkullorJade, ⁠⁠⁠⁠⁠⁠⁠@Miss_Magitek⁠⁠⁠⁠⁠⁠⁠ and ⁠⁠⁠⁠⁠⁠⁠@Binary_Dragon⁠⁠⁠⁠⁠⁠⁠,⁠⁠⁠⁠⁠⁠⁠ @retrodatv⁠⁠⁠⁠⁠⁠⁠ on twitch, for live D&D, TTRPGs and more.#cyberpunkred #actualplay #ttrpg #radioplay #scifi #cyberpunk #drama #comedy #LIVE #SDCC25

How do we know through atmospheres? How can being affected by an atmosphere give rise to knowledge? What role does somatic, nonverbal knowledge play in how we belong to places? Atmospheric Knowledge takes up these questions through detailed analyses of practices that generate atmospheres and in which knowledge emerges through visceral intermingling with atmospheres. From combined musicological and anthropological perspectives, Birgit Abels and Patrick Eisenlohr investigate atmospheres as a compelling alternative to better-known analytics of affect by way of performative and sonic practices across a range of ethnographic settings. With particular focus on oceanic relations and sonic affectedness, Atmospheric Knowledge centers the rich affordances of sonic connections for knowing our environments. A free ebook version of this title is available through Luminos, University of California Press's Open Access publishing program. Visit www.luminosoa.org to learn more. Learn more about your ad choices. Visit megaphone.fm/adchoices Support our show by becoming a premium member! https://newbooksnetwork.supportingcast.fm/new-books-network

How do we know through atmospheres? How can being affected by an atmosphere give rise to knowledge? What role does somatic, nonverbal knowledge play in how we belong to places? Atmospheric Knowledge takes up these questions through detailed analyses of practices that generate atmospheres and in which knowledge emerges through visceral intermingling with atmospheres. From combined musicological and anthropological perspectives, Birgit Abels and Patrick Eisenlohr investigate atmospheres as a compelling alternative to better-known analytics of affect by way of performative and sonic practices across a range of ethnographic settings. With particular focus on oceanic relations and sonic affectedness, Atmospheric Knowledge centers the rich affordances of sonic connections for knowing our environments. A free ebook version of this title is available through Luminos, University of California Press's Open Access publishing program. Visit www.luminosoa.org to learn more. Learn more about your ad choices. Visit megaphone.fm/adchoices Support our show by becoming a premium member! https://newbooksnetwork.supportingcast.fm/anthropology

How do we know through atmospheres? How can being affected by an atmosphere give rise to knowledge? What role does somatic, nonverbal knowledge play in how we belong to places? Atmospheric Knowledge takes up these questions through detailed analyses of practices that generate atmospheres and in which knowledge emerges through visceral intermingling with atmospheres. From combined musicological and anthropological perspectives, Birgit Abels and Patrick Eisenlohr investigate atmospheres as a compelling alternative to better-known analytics of affect by way of performative and sonic practices across a range of ethnographic settings. With particular focus on oceanic relations and sonic affectedness, Atmospheric Knowledge centers the rich affordances of sonic connections for knowing our environments. A free ebook version of this title is available through Luminos, University of California Press's Open Access publishing program. Visit www.luminosoa.org to learn more. Learn more about your ad choices. Visit megaphone.fm/adchoices Support our show by becoming a premium member! https://newbooksnetwork.supportingcast.fm/sociology

How do we know through atmospheres? How can being affected by an atmosphere give rise to knowledge? What role does somatic, nonverbal knowledge play in how we belong to places? Atmospheric Knowledge takes up these questions through detailed analyses of practices that generate atmospheres and in which knowledge emerges through visceral intermingling with atmospheres. From combined musicological and anthropological perspectives, Birgit Abels and Patrick Eisenlohr investigate atmospheres as a compelling alternative to better-known analytics of affect by way of performative and sonic practices across a range of ethnographic settings. With particular focus on oceanic relations and sonic affectedness, Atmospheric Knowledge centers the rich affordances of sonic connections for knowing our environments. A free ebook version of this title is available through Luminos, University of California Press's Open Access publishing program. Visit www.luminosoa.org to learn more. Learn more about your ad choices. Visit megaphone.fm/adchoices Support our show by becoming a premium member! https://newbooksnetwork.supportingcast.fm/geography

How do we know through atmospheres? How can being affected by an atmosphere give rise to knowledge? What role does somatic, nonverbal knowledge play in how we belong to places? Atmospheric Knowledge takes up these questions through detailed analyses of practices that generate atmospheres and in which knowledge emerges through visceral intermingling with atmospheres. From combined musicological and anthropological perspectives, Birgit Abels and Patrick Eisenlohr investigate atmospheres as a compelling alternative to better-known analytics of affect by way of performative and sonic practices across a range of ethnographic settings. With particular focus on oceanic relations and sonic affectedness, Atmospheric Knowledge centers the rich affordances of sonic connections for knowing our environments. A free ebook version of this title is available through Luminos, University of California Press's Open Access publishing program. Visit www.luminosoa.org to learn more. Learn more about your ad choices. Visit megaphone.fm/adchoices Support our show by becoming a premium member! https://newbooksnetwork.supportingcast.fm/sound-studies

The Crew and some new and old friends, take on the Mass Driver at Morro Rock. Domino has discovered a plot to take out the crew's communication satellite and time is ticking before it's blown out of the sky.Jade, Evan and Peter are joined by Dayeanne Hutton and Anais R Morgan (Infinite Sided Dice) live on stage at San Diego Comic Con 2025!Check out our Youtube Channel for more live panels! HereBTS and art posts will come out for all our patrons to peruse, after Part 2 is released next week!More info can be found here: ⁠⁠⁠⁠⁠⁠linktr.ee/NoLatency⁠⁠⁠⁠⁠⁠If you'd like to support us, We now have a Patreon! ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Patreon.com/nolatency⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Even more information and MERCH is on our website! ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠www.nolatencypodcast.com⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Twitter: @nolatencypodInstagram: @nolatencypodFind @SkullorJade, ⁠⁠⁠⁠⁠⁠⁠@Miss_Magitek⁠⁠⁠⁠⁠⁠⁠ and ⁠⁠⁠⁠⁠⁠⁠@Binary_Dragon⁠⁠⁠⁠⁠⁠⁠,⁠⁠⁠⁠⁠⁠⁠ @retrodatv⁠⁠⁠⁠⁠⁠⁠ on twitch, for live D&D, TTRPGs and more.#cyberpunkred #actualplay #ttrpg #radioplay #scifi #cyberpunk #drama #comedy #LIVE #SDCC25

In this week's vBrownBag, Principal Software Engineer Dominik Wosiński takes us on a deep dive into Amazon Nova Sonic — AWS's latest speech-to-speech AI model. Dominik explores how unified voice models like Nova Sonic are reshaping customer experience, DevOps workflows, and real-time AI interaction, with live demos showing just how natural machine-generated speech can sound. We cover what makes speech-to-speech difficult, how latency and turn-detection affect conversational design, and why this technology marks the next frontier for AI-driven customer support. Stick around for audience Q&A, live experiments, and insights on where AWS Bedrock and generative AI are headed next.

Bill Severn of 1623 Farnam joins JSA TV from DCD>Connect Virginia to discuss how #GenerativeAI is reshaping network infrastructure. He shares insights on latency limits, #edge inference, #hyperscaler-driven metro upgrades, and what an AI-ready interconnect looks like. Plus, a look ahead at 1623 Farnam's expansion plans and investments for the next 12–24 months.

“AI is hungry — for bandwidth, for speed, and for talent.” — Jean-Philippe Avelange, Chief Information Officer, Expereo Jean-Philippe Avelange, CIO of Expereo, joined Doug Green, Publisher of Technology Reseller News, to discuss findings from Expereo's Horizon Telecom Report—revealing how U.S. organizations are losing millions to network failures and struggling to find skilled professionals in cybersecurity, networking, and data automation. Avelange explained that as companies digitize everything from collaboration to customer experience, connectivity interruptions now directly halt business operations, making network reliability as vital as cybersecurity. “Modern enterprises are building their products and services on connectivity. When it stops, business stops,” he noted. The AI multiplier AI adoption is compounding the challenge. “AI is not just another workload—it's a new kind of demand,” Avelange said. AI-driven automation, real-time data flows, and low-latency interactions place unprecedented pressure on legacy network architectures. Organizations can no longer treat networking as a commodity; they must rethink it as a strategic platform requiring redesign and intelligent automation. The human factor According to Avelange, the real shortage isn't people—it's adaptability. The industry needs professionals skilled in network automation, data flow optimization, and problem solving, not just hardware management. “AI won't solve your problem if you don't understand the problem,” he said, advocating for upskilling internal teams alongside strong partnerships with managed service providers (MSPs) that bring intelligence, not just infrastructure. Latency by design Latency, Avelange warned, must be addressed before deployment. “You can always add bandwidth, but you can't add speed after the fact. Latency has to be engineered from the start.” A new mindset For Expereo, the future of networking lies in intelligent connectivity—solutions that merge automation, analytics, and agility to keep enterprises resilient in the AI era. “We're not selling boxes,” Avelange said. “We're helping companies design the networks their digital business runs on.” Read more in the Horizon Telecom Report or visit expereo.com.

In this Mission Matters session hosted by ⁠Adam Torres⁠, ⁠Eraj Akhtar⁠ (CTO & Co-Founder, Excite Capital LLC) and ⁠Namuun Battulga⁠ (CEO, Jenko Tour JSC & Igo Hotel and Resorts) discuss physics-based, quantum-inspired AI trading and Mongolia's emergence as a cost-efficient, secure data center location powered by a new 70MW plant. They share partner criteria, address security considerations, and outline a mission to scale globally distributed compute and real-economy growth across Asia. Follow Adam on Instagram at ⁠https://www.instagram.com/askadamtorres/⁠ for up to date information on book releases and tour schedule. Apply to be a guest on our podcast: ⁠https://missionmatters.lpages.co/podcastguest/⁠ Visit our website: ⁠https://missionmatters.com/⁠ More FREE content from Mission Matters here: ⁠https://linktr.ee/missionmattersmedia⁠ Learn more about your ad choices. Visit podcastchoices.com/adchoices

In this Mission Matters session hosted by Adam Torres, Eraj Akhtar (CTO & Co-Founder, Excite Capital LLC) and Namuun Battulga (CEO, Jenko Tour JSC & Igo Hotel and Resorts) discuss physics-based, quantum-inspired AI trading and Mongolia's emergence as a cost-efficient, secure data center location powered by a new 70MW plant. They share partner criteria, address security considerations, and outline a mission to scale globally distributed compute and real-economy growth across Asia. Follow Adam on Instagram at https://www.instagram.com/askadamtorres/ for up to date information on book releases and tour schedule. Apply to be a guest on our podcast: https://missionmatters.lpages.co/podcastguest/ Visit our website: https://missionmatters.com/ More FREE content from Mission Matters here: https://linktr.ee/missionmattersmedia Learn more about your ad choices. Visit podcastchoices.com/adchoices

Join Alex Golding as he sits down with Austin Federa, Co-founder of DoubleZero, to explore how they're building permissionless high-performance fiber infrastructure that could revolutionize blockchain performance. Austin shares the technical vision behind creating a parallel internet for distributed systems, starting with Solana validators as their initial market.DoubleZero: https://doublezero.xyz

Jimmy Bogard joins Pod Rocket to talk about making monoliths more modular, why boundaries matter, and how to avoid turning systems into distributed monoliths. From refactoring techniques and database migrations at scale to lessons from Stripe and WordPress, he shares practical ways to balance architecture choices. We also explore how tools like Claude and Lambda fit into modern development and what teams should watch for with latency, transactions, and growing complexity. Links Website: https://www.jimmybogard.com X: https://x.com/jbogard Github: https://github.com/jbogard LinkedIn: https://www.linkedin.com/in/jimmybogard/ Resources Modularizing the Monolith - Jimmy Bogard - NDC Oslo 2024: https://www.youtube.com/watch?v=fc6_NtD9soI Chapters We want to hear from you! How did you find us? Did you see us on Twitter? In a newsletter? Or maybe we were recommended by a friend? Fill out our listener survey (https://t.co/oKVAEXipxu)! Let us know by sending an email to our producer, Em, at emily.kochanek@logrocket.com (mailto:emily.kochanek@logrocket.com), or tweet at us at PodRocketPod (https://twitter.com/PodRocketpod). Follow us. Get free stickers. Follow us on Apple Podcasts, fill out this form (https://podrocket.logrocket.com/get-podrocket-stickers), and we'll send you free PodRocket stickers! What does LogRocket do? LogRocket provides AI-first session replay and analytics that surfaces the UX and technical issues impacting user experiences. Start understanding where your users are struggling by trying it for free at LogRocket.com. Try LogRocket for free today. (https://logrocket.com/signup/?pdr) Special Guest: Jimmy Bogard.

Monitoring and troubleshooting latency can be tricky. If it’s in the network, was it the IP stack? A NIC? A switch buffer? A middlebox somewhere on the WAN? If it’s the application, can you, the network engineer, bring receipts to the app team? And what if you need to build and operate a network that’s... Read more »

Monitoring and troubleshooting latency can be tricky. If it’s in the network, was it the IP stack? A NIC? A switch buffer? A middlebox somewhere on the WAN? If it’s the application, can you, the network engineer, bring receipts to the app team? And what if you need to build and operate a network that’s... Read more »