Podcasts about Expense

Asking for the bill is one of the most important things to learn in any language. But חשבון, in Hebrew, is about much more than just settling the account. Guy explains how Israelis do self-reflection, how they break even… and how they don't give a damn. Hear the All-Hebrew Episode on Patreon   New Words and Expressions: Heshbon, heshbonot (m.) – Bill, bills – חשבון, חשבונות Heshbon bank – Bank account – חשבון בנק Heshbon hotsa'ot – Expense account – חשבון הוצאות Heshbon nefesh – Self examination – חשבון נפש Lehavi / Lakachat be-heshbon – To take into consideration – להביא / לקחת בחשבון Kach / K'chi / K'choo be-heshbon she- – Take into consideration that… (Imp.) – קח / קחי / קחו בחשבון K'chi be-heshbon she-kar – Take into consideration that it's cold – קחי בחשבון שקר Kach be-heshbon she-yakar sham – Take into consideration that it's expensive there – קח בחשבון שיקר שם Lo ba be-heshbon – It's out of the question – לא בא בחשבון "Panim she-lo osot heshbon" – A face that does not give a damn – פנים שלא עושות חשבון Hu lo dofek heshbon – He doesn't give a damn – הוא לא דופק חשבון Ma, ata dofek heshbon le-mishehu? – Do you care about others? – מה, אתה דופק חשבון למישהו Hi sogeret heshbonot – She's clearing the table – היא סוגרת חשבונות Hisool heshbonot – Score settling – חיסול חשבונות Zeh al heshboncha? – Is it at your expense – זה על חשבונך Al heshbon ha-bayit – On the house – על חשבון הבית Zeh al heshbonenu – This is on our account – זה על חשבוננו Heshbon aroch – A long-standing score – חשבון ארוך Yesh li heshbon aroch ito – I have a long-standing score to settle with him – יש לי חשבון ארוך איתו Be-heshbon pashut – In simple arithmetic – בחשבון פשוט Ro'eh heshbon – Accountant – רואה חשבון Heshbona'ut / Re'iyat heshbon – Accounting – חשבונאות / ראיית חשבון Hanhalat heshbonot – Bookkeeping – הנהלת חשבונות Menahel / menahelet heshbonot – Bookkeeper – מנהל / מנהלת חשבונות Tavi'i cheshbonit – Bring an invoice (imp. f.) – תביאי חשבונית Heshbonit mas – Tax invoice – חשבונית מס Lehashben le-mishehu – To care too much about something – לחשבן למישהו Lehitchashben – To settle accounts with someone – להתחשבן Bo nitchashben ba-sof – Let's do the math at the end – בוא נתחשבן בסוף Ma, ata mitchashben iti al cafe? – Forget it, it's just a coffee, my treat! – מה, אתה מתחשבן איתי על קפה Hitchashbenut – Settling an account – התחשבנות    Playlist and Clips: Ariel Zilber – Holech Batel (lyrics) Ofra Haza – Shir Ha-frecha (lyrics) Rami Kleinstein & Ha-mo'atsa – Ha-boker At Holechet (lyrics) Tea Packs & Alma Zak – Perech Ha-shchunot (lyrics) Ep. 172 about to knock Ep. 440 about receipt and invoice HEB

Have you ever caught yourself saying, "I'm probably just overthinking it"?That's exactly what I said in a recent supervision session. My supervisor responded with a simple question:"Are you overthinking - or are you diminishing your feelings?"That question stayed with me.As therapists, we're often deeply compassionate people. We want to be kind, fair and understanding. But sometimes our kindness towards other people comes at the expense of ourselves.In this episode, I reflect on how easy it is to ignore our own needs because we're worried about disappointing someone else. Whether it's charging a cancellation fee, ending a session on time, saying no to a new client or making a difficult professional decision, our fear of disapproval can sometimes become louder than our own values.I also explore why consistency matters, how resentment can grow when we don't honour our own boundaries, and why listening to ourselves is an important part of ethical, sustainable practice.In this episode, we explore:Why "overthinking" isn't always what we're doingThe difference between kindness and self-neglectHow people pleasing can influence our professional decisionsWhy your values are a better guide than someone else's approvalHow healthy boundaries protect both you and your clientsThe importance of looking after yourself as well as those you supportIf you've ever struggled with saying no, worried about getting things wrong or found yourself putting everyone else's needs before your own, I hope this episode reminds you that your wellbeing matters too.If you'd like support from a community of therapists who understand these challenges, Therapy Growth Group offers weekly live calls, practical marketing support and a welcoming place to think things through with others in private practice.Setting up in private practice? Download my free checklist HERENeed ideas for how to get clients? Download my free handout 21 Ways for Counsellors to Attract New Clients HEREYou can also find me here:The Good Enough Counsellors Facebook GroupJosephine Hughes on FacebookJosephine Hughes on YouTubeMy website: josephinehughes.comKeywords: therapists, counsellors, psychotherapy, private practice, therapist boundaries, people pleasing, self care, burnout prevention, compassion fatigue, professional boundaries, supervision, therapist wellbeing, values, cancellation policy, counselling podcastThe information contained in Good Enough Counsellors is provided for information purposes only. The contents of this podcast are not intended to amount to advice and you should not rely on any of the contents of this podcast. Professional advice should be obtained before taking or refraining from taking any action as a result of the contents of this podcast.Josephine Hughes disclaims all liability and responsibility arising from any reliance placed on any of the contents of this podcast.

U.S. Representative Dan Newhouse says with the labor shortages and high expenses, there is a lot of misinterpretations of the ag labor force in general.

Most eCom marketers are optimizing for the wrong number, and a real CFO explains exactly how to fix that! Our good friend Abir Syed has a rare take on how to scale brands: he's run an eCommerce brand, built a performance marketing agency, and now runs a fractional CFO firm specializing in eCom. In this episode, he breaks down why MER is basically useless, what cohort profit actually tells you, and the three-pillar finance framework that gives marketers and CFOs a shared language for growth. Nate also gets uncomfortably personal about his own brand's cash flow situation. You'll walk away understanding how to set real scaling targets, why over-revving your marketing engine costs you money, how to predict LTV decay as you shift from organic to paid customers, and what questions to actually ask your finance team. 00:00 Why selling out early isn't the flex you think it is01:45 Introducing Abir Syed — the CFO who hates accounting03:30 Why 80% of your business story lives in the finances05:30 The #1 thing to get right before anything else: inventory costing07:00 Why MER is a useless metric (hot take, but hear him out)08:30 How dropping MER led to 130% growth in one year09:15 Cohort profit: Abir's favorite metric explained13:00 Over-revving the engine — the hidden way brands lose money scaling16:30 Incrementality testing vs. the scaling target table18:00 LTV decay: what happens when you go from organic to paid acquisition20:00 Paid customers get stolen by ads — a concept that breaks your brain23:00 The 3 things a CFO and CMO should never fight about24:30 Pillar 2: Investing in the marketing engine (creative, tools, talent)26:00 Pillar 3: Cash flow strategy and payback periods33:30 Expense leverage — making sure every dollar has a job37:00 Nate's regret: not taking big shots during a 2.5-year hot streak38:30 The $100K YouTube deal that looked like a disaster until Q439:30 The one thing to do this week if you're looking at your numbers

EDITORIAL: SEC over-regulation at the expense of shareholder rights | Aug. 13, 2026Check out our Streaming Channel: https://streaming.manilatimes.net/Subscribe to The Manila Times Channel - https://tmt.ph/YTSubscribeVisit our website at https://www.manilatimes.netFollow us:Facebook - https://tmt.ph/facebookInstagram - https://tmt.ph/instagramTwitter - https://tmt.ph/twitterDailyMotion - https://tmt.ph/dailymotionSubscribe to our Digital Edition - https://tmt.ph/digitalCheck out our Podcasts:Spotify - https://tmt.ph/spotifyApple Podcasts - https://tmt.ph/applepodcastsAmazon Music - https://tmt.ph/amazonmusicDeezer: https://tmt.ph/deezerStitcher: https://tmt.ph/stitcherTune In: https://tmt.ph/tunein#TheManilaTimes#VoiceOfTheTimes Hosted on Acast. See acast.com/privacy for more information.

Government often relies on industry to invest ahead of future requirements. The recent pause in CMMC implementation is raising questions about how contractors weigh risk, cost and the value of moving first. Here with one company's perspective are Angie Lienert and Jeremiah Jensen of IntelliGenesis.See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

Paul returns from three days at the Garrett Planning Network retreat with a lesson that has almost nothing to do with investments — and everything to do with getting your money's worth from professional advice.Garrett advisors work by the hour, a business model Paul believes eliminates the conflicts of interest built into assets-under-management relationships. For $1,000 to $8,000, he's convinced most families can get extraordinary value from five to ten hours with a thoughtful, trained hourly planner. But there's a catch: the value of those hours depends almost entirely on your willingness to tell the truth. Inspired by a Seth Godin observation — people lie in focus groups, on surveys, and to themselves — Paul explains why the most valuable planning meeting isn't the one where you look financially successful. It's the one where you're completely honest. Paul and his wife are putting this to the test with an hourly planner of their own, and he'll report back in the weeks ahead.Next, Paul shares a private conversation with his longtime friend Rick Ferri, who challenged an idea Paul has taught for decades: that small cap value, large cap value, and international are equity asset classes at all. Rick argues there's only one equity asset class — the total market — and everything else is a segment or style. Paul takes the challenge seriously, does some digging, and explains why the answer matters far more than a debate over definitions. How you think about asset classes shapes the portfolio you'll live with for the next 60 or 70 years.Finally, Paul digs into AVGE, the Avantis globally diversified all-equity ETF, and how it compares to Vanguard's total market approach (VT and VTI). He walks through the meaningful differences: 70/30 U.S./international at Avantis versus 60/40 at Vanguard, and substantially larger positions in mid cap value, small cap value, and small cap blend. He looks at what those tilts have meant historically — including Vanguard's own mid cap value fund turning $10,000 into roughly $160,000 versus $102,000 for the S&P 500 — and why he believes the extra 0.17% in expenses may be money well spent. For investors who don't want to go all-in, Paul offers simple combinations, like a third VT, a third AVGE, and a third AVUV.CHAPTERS00:00 – Introduction: three topics from the Garrett retreat01:56 – Why hourly advisors have fewer conflicts of interest05:52 – The catch: your willingness to tell the truth06:38 – Seth Godin: "People lie... and they lie to themselves"08:04 – What planners can't fix if they don't know about it13:00 – Paul's debate with Rick Ferri: what is an equity asset class?18:05 – Why the definition shapes your lifetime portfolio21:34 – AVGE vs. VT: U.S./international balance23:07 – Comparing value, blend, and growth exposure25:00 – Mid cap and small cap: what history shows30:15 – Expense ratios and what you're paying for31:35 – Simple combinations: VT + AVGE + AVUV33:15 – Stay the course: closing thoughtsLearn more about the Garrett Planning Network

Teaching Kids How To Spot Manipulation (but at the family's expense?) Are we protecting our kids from family drama, or are we missing crucial teachable moments about manipulation and boundaries?

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

Who is responsible for deciding whether a business expense is tax deductible: the client, the bookkeeper, or the tax preparer? In this episode, our favorite Bookkeeping Mensch, Paul Rosenblum, shares his approach to handling missing receipts, client gifts, business meals, and other bookkeeping gray areas. He touches on where professional responsibility begins and ends, and whether bookkeepers should rely on client information or use their own judgment when categorizing expenses. Send us Fan MailSupport the showAbout the hostPaul Rosenblum has been doing hands-on bookkeeping for over 30 years, starting with QuickBooks Desktop and adapting to the world of cloud-based QuickBooks Online. He shares practical, in-the-weeds lessons from real client files every episode.

We all know life feels more expensive, but one everyday household expense has skyrocketed by a staggering 200% over the past decade. Webpage

Trudie Mason is joined by Meeker Guerrier, Commentator at Noovo and RDS, and Andrew Caddell, columnist for the Hill Times and President of the Task Force on Linguistic policy. The premiers of nine provinces announced that they’re removing major barriers to interprovincial sales of alcohol. While Quebec Premier Christine Fréchette agreed with the deal, she was not one of the premiers who signed it. A new Leger poll commissioned by the Canadian Taxpayers Association says that over 70 percent of Canadians want to end the expense accounts of former governors general. After being pressured for the past few weeks, the federal government has released details of its revised agreement with the United States on the Gordie Howe International Bridge. A new Leger report says 75 per cent of Canadians think AI use should be disclosed for schoolwork and 61 per cent for work. Two columnists with La Presse last week wrote a dialogue piece exploring whether Montreal is becoming harder to love.

I can't believe what we spent during the first half of 2026!? Our expenses have definitely added up and even surprised us, which is why we think tracking them is so important. We lay out all of our mid-year expenses and discuss unexpected life events, house repairs, and intentional shifts in daily habits that affect everyone's expenses. This is a fully transparent look at our real-life expenses and how we feel about them. Get the full show notes, show references, and more information here: https://www.insideoutmoney.org/172-mid-year-2026-expense-review-broken-bones-new-roofs-and-130-airport-meals/

The UV technology has been purchased, delivered, and rolled into the facility. So now what? In the final episode of "Evidence Over Expense," Kyle Morrison and Xenex's Selena Pfannstiel join us to explore why the real return on investment begins after the sale is complete. From staff training and user adoption to reporting, accountability, workflow integration, and ongoing vendor support, this conversation breaks down the often-overlooked factors that can make or break long-term program success. If your organization is investing in UV technology to reduce transmission risk and drive lasting value, this episode will change the way you think about what happens after the rollout. Tune in to this series finale today! Once you complete the interview, jump on over to the link below to take a short quiz and download your CEC certificate for 0.5 CECs! – https://www.flexiquiz.com/SC/N/ps-xenex-ep3 A special thanks to our sponsor, Xenex, for making this series possible. #PowerSupply #Xenex #HealthcareSupplyChain #InfectionPrevention #ValueAnalysis #EvidenceOverExpense #UVTechnology #Podcast

LISTEN and SUBSCRIBE on:Apple Podcasts: https://podcasts.apple.com/us/podcast/watchdog-on-wall-street-with-chris-markowski/id570687608 Spotify: https://open.spotify.com/show/2PtgPvJvqc2gkpGIkNMR5i WATCH and SUBSCRIBE on:https://www.youtube.com/@WatchdogOnWallstreet/featured  Wall Street's biggest firms are posting record trading revenue while amateur investors chase quick profits through options and addictive trading platforms. The house always wins—and individual traders are competing against sophisticated algorithms, superior information, and firms capable of manipulating every market move. Wall Street is having its best year ever because millions of inexperienced investors still believe they can beat the professionals at their own game.

Bingo: https://bingobaker.com#6a2f5f147a86d197

In this podcast, Jandevman gives us an overview of what they do at Liquidy, how the project originated, and how they ended up building on THORChain.Swap now https://swap.thorchain.org/THORChain is a decentralized crypto exchange. THORChain is the first and biggest DEX for Bitcoin. You can use any self custody wallet to swap and there's no KYC required.Timestamps:00:00:00 Intro00:02:00 Kenton update — Keplr should be working! More integrations are coming!00:03:00 Affiliate page widget is ready for testing00:04:00 Air Canada story00:06:00 Jandevman introduction00:10:00 What was so different about Kujira that attracted you to it?00:13:00 Jandevman rebuilt everything from scratch00:16:00 Treasury discussion — what were the assets?00:17:00 What was MantaDAO?00:19:00 Website walkthrough00:21:00 Valuation breakdown — comparing Liquidy to TradFi00:23:00 Kenton explains NAV and asks about redemptions00:24:00 Financial reports and analytics breakdown00:26:00 Expense breakdown00:27:00 Revenue sources00:29:00 Governance process breakdown00:30:00 Simple majority or supermajority?00:31:00 Veto power through a multisig00:32:00 Bonk comparison00:36:00 Raising external capital?00:38:00 Growing the treasury00:39:00 How does market making work?00:43:00 Why isn't the LQDY token on the base layer?00:45:00 How many wallets support secured assets?00:47:00 Which other wallets should support secured assets?00:48:00 What other financial primitives will LQDY be involved in?00:52:00 Will LQDY get involved in lending?00:55:00 What other assets will be acquired?00:56:00 What about Auto Rujira?00:57:00 Liquidity swap router breakdown00:59:00 What's the difference between the swap router on Rujira and LQDY?01:02:00 Comparing different liquidity pools01:07:00 Will redacted functionality be available?01:09:00 bRUNE breakdown01:11:00 What is a good APY for users?01:13:00 Will STO use the LQDY API?01:15:00 Is there anything else you'd like to add?01:18:00 Transparency discussion01:19:00 Proposal discussion01:22:00 Everything is on-chain01:27:00 Thoughts on inflation01:30:00 The model is beautifully simple01:33:00 We need to do this again!01:34:00 Conclusion

Guest host Rob Fai & Marit Stiles, MPP for Davenport, leader of the Ontario NDP and the leader of the Official Opposition discuss: 1 - Ford breaks silence on hotel expense scandal 2 - Ford's office freezes hiring, eliminates positions to cut costs, memo shows 3 - Wildfires threaten Northern Ontario communities as province asks for federal help 4 - Gordie Howe Bridge's net revenue to be ‘modest' in early years, Mark Carney says Learn more about your ad choices. Visit megaphone.fm/adchoices

Guest host Brad Fai & Colin D'Mello, Global News Queen's Park Bureau Chief discuss: 1 - ‘They're paying back the money': Ford breaks silence on hotel expense scandal 2 - Ford's office reduces staff costs by $1M following spending criticism 3 - Ford defends Ontario emergency firefighting budget as province asks Ottawa to help with evacuations Learn more about your ad choices. Visit megaphone.fm/adchoices

Every dollar sitting in your bank account is quietly losing value, and most people never realize why. On this episode of Play Big Faster, former $10 billion portfolio manager Paul Musson, author of Capital Offense: Why Some Benefit at Your Expense, breaks down the hidden inflation tax draining entrepreneurial wealth. Paul unpacks how does inflation affect your savings, why the real inflation rate differs from official numbers, and how does the government undercount inflation through substitution and hedonic adjustments. You will hear what happens to savings during inflation, why does printing money cause inflation, whether inflation hurts the middle class, and steps toward inflation proof wealth. Ideal for founders who want to protect what they have built while still growing. Tune in for straight talk on why your cost of living keeps going up.

JAY TRUITT TALKS FEED MILL FIRES AND THE FIGHT OVER AI DATA CENTERS ON FARMLAND Trent Loos welcomes Jay Truitt from Texas for a hard-hitting look at issues shaking rural America. The two dig into a troubling rise in feed mill fires, breaking down decades of data and the real dangers farmers and…

Guest host Rob Fai spoke with Colin D'Mello Global News Queen's Park Bureau Chief about Ford government minister bills taxpayers $16K for Toronto hotels despite living in city Learn more about your ad choices. Visit megaphone.fm/adchoices

Keith Weinhold explains why inflation has become a permanent part of the post–World War II economy and what that shift means for today's financial system.  He breaks down economist Dr. Mark Skousen's five structural reasons behind never-ending inflation and ties them to the hollowing out of the middle class and the "last generation to live normally" concept.  Keith then introduces opportunity cost as the biggest financial expense most people overlook and illustrates how leveraging low-cost, long-term debt to buy productive real assets can turn inflation into an advantage.  He closes by outlining a practical hierarchy for which debts to eliminate first and which to keep as tools for long-term wealth building. Episode Page: GetRichEducation.com/614 For access to properties or free help with a GRE Investment Coach, start here: GREmarketplace.com GRE Free Investment Coaching: GREinvestmentcoach.com Get mortgage loans for investment property: RidgeLendingGroup.com or call 855-74-RIDGE  or e-mail: info@RidgeLendingGroup.com Invest with Freedom Family Investments.  For predictable 10-12% quarterly returns, visit FreedomFamilyInvestments.com/GRE or text  FAMILY to 66866  Unlock truly passive real estate income—visit flockhomes.com/GRE today to see if your properties qualify for a 721 exchange with Flock Homes. To get in the best physical, mental, and professional shape of your life, go to DanielThomasHind.com and apply for Daniel's intensive 1-on-1 coaching for burnt-out entrepreneurs and executives. Will you please leave a review for the show? I'd be grateful. Search "how to leave an Apple Podcasts review"  For advertising inquiries, visit: GetRichEducation.com/ad Best Financial Education: GetRichEducation.com Get our wealth-building newsletter free— GREletter.com  Our YouTube Channel: www.youtube.com/c/GetRichEducation Follow us on Instagram: @getricheducation Complete episode transcript:   Keith Weinhold  0:01   Welcome to GRE. I'm your host Keith Weinhold. In less than 40 years, America has gone from 75% gasoline to permanent inflation. Then learn about the biggest financial expense you will ever have in your life. It's not taxes, housing, interest charges, inflation, children, or healthcare. Most people have never heard of it today on Get Rich Education. You know, Mid South Homebuyers, that top Memphis turnkey provider. I learned that a secret weapon behind their explosive growth is more than just you buying their properties. It's an executive coach. For nine years now. Their CEO Terry Kerr and his COO Pat Nix have worked privately with a coach who I've now learned from too, and he doesn't market himself online anywhere. After 12 years behind the scenes, that coach is now making himself available exclusively for GRE listeners. His name is Daniel Thomas Hind. If you're a hard-charging business owner or investor who wants to get in the best shape of your life, physically, mentally, and professionally, you can fill out an application for a free consult. This is private one-on-one coaching for those willing to go to uncommon lengths to achieve uncommon results. Thanks to Daniel, we've all become better leaders, better operators, and better men. It started by showing up for ourselves. Now it's your turn. Go to DanielThomashHind.com. H-I-N-D. That's DanielThomashHind.com, and sign up before spots fill.   Keith Weinhold  1:41   What if you got your mortgage loans the same place I get mine? You sure can at Ridge Lending Group NMLS 42056 They provided GRE listeners with more loans than anyone because Ridge specializes in investment property. They'll help you build a long-term plan for growing your real estate empire with leverage. Start your prequal and even chat directly with President Chaley Ridge. While it's on your mind, start at ridgelendinggroup.com. That's ridgelendinggroup.com.   Speaker 1  2:14   You're listening to the show that has created more financial freedom than nearly any show in the world, this is Get Rich Education.   Keith Weinhold  2:31   Welcome to GRE from Bavaria, Germany, to Batavia, New York, and across 188 world nations. I'm Keith Weinhold, and you're listening to Get Rich Education. In the 19 the 1988 movie Die Hard, there's a California gas station sign in the background that's visible. You can see it there. The gas price on this sign is a jaw dropper. Unleaded 77.9 cents per gallon, regular 70-4.9 cents per gallon. That now looks like it belongs in a museum next to rotary phones and blockbuster video cards. Yes, California gas for 75 cents, and the movie Die Hard. It had all these actors from yesteryear, like Bruce Willis and Reginald Vel Johnson. Yet you, depending on your age, you might remember 1988. It's not like ancient history. Now we all know that inflation is always and everywhere a monetary phenomenon, like Milton Friedman said, but is there more to this? Is there more than the Fed targeting 2% inflation, just like it says on their website? Oh, there sure is. And by the way, with a little research, it looks like California Gas averaged 95 cents in 1988, not 75 like it shows in Die Hard, but in any case, the point is still there. And today, inflation keeps running hot. Four years ago, the pandemic made CPI inflation peak at 9.1 percent. Today, the hangover effects of tariffs push it up, and the Iran war are turning up the heat even more, with the latest reading above 4% Inflation is running at more than double what the Fed wants. You can even make the case now that inflation is out of control. But here's the thing: inflation has exceeded that 2% target for 60-three consecutive months now. I mean, think about what that means. My gosh, just imagine having an important target that affects every American and missing it 60-three times in a row. That's kind of what's happening now, and they're. Going to keep missing it. So this streak of inflation above 2% started back in March of 2021 during the pandemic hangover, and it is still going strong after 63 months. Nobody knows where this is going to end. Most Americans get crushed by rising prices because their wages don't keep up, and you know collectively they sort of think we are concerned, but then they mostly keep doing the same thing while their lifestyle quietly shrinks. So consumers despise inflation. Everyday investors are lukewarm about inflation, and leverage real estate investors are smiling like they found a 20-dollar bill in last winter's coat. Leverage real estate investors are pretty ecstatic about inflation. Now the history gets super interesting.   Keith Weinhold  5:59   Okay, how did we get into this, where we just always seem to have inflation? So learn the history, and then I'll tie it back to how it affects you as an investor. Because before World War II, inflation behaved differently. The old pre-1945 pattern was that we had inflation during wars and booms. We had deflation after panics and depressions. So therefore, the result was that over long stretches, price levels often just moved sideways. We used to have recessions more often back 80 plus years ago than we do now. So therefore, you just had these price levels move sideways because a recession even prompted deflation, actually a strengthening of purchasing power. But then after World War II, inflation basically went permanently positive. I mean, yeah, permanently positive, where inflation is just always turned on with very few exceptions to that. In wartime, now we have inflation. In peacetime, now we have inflation. During the Super Bowl, now we have inflation. It is inflation, no matter what is going on. Right then, so what changed? Prominent economist and GRE podcast guest here, Dr. Mark Skousen. He has cited five major reasons that inflation became a permanent fixture from 1945 until today. And Mark Skousen was here on the show with us almost exactly two years ago because he's also the founder of a great event called Freedom Fest that Nareesh and I broadcast a show from, the five reasons that Scowson cites for never-ending inflation are first, never-ending wars. Now this doesn't only mean formally declared boots on the ground wars where tanks are rolling, never-ending wars. It means this permanent state of global military readiness that we have today, where we have overseas bases, defense contractors, right with the military-industrial complex. We have NATO commitments.   Keith Weinhold  8:17   We have anti-terror operations, naval patrols, intelligence agencies, and all this enormous machinery that's required to keep America as the world's security backstop. Well, all that costs an awful lot of money, and when government wants more money than it collects, it has a favorite trick: just create more dollars and create them out of nothing. I mean, it's like ordering another round of drinks for the table and then putting it on the unborn grandchildren's tab. The second reason for the never-ending inflation is the 1913 creation of the Federal Reserve and how that's changed over time because the Fed they were originally supposed to defend the dollar, defend the gold standard, and act as lender of last resort. Today it mostly just does the last one. It acts as the lender of last resort, and it's really not even last resort. I mean, she shit seems to patch any significant hole in the economy by creating more dollars and then pumping them into the system. When markets wobble, banks panic, or politicians overspend, or the economy catches any kind of cold, you know, the Fed often just shows up with this fire hose of liquidity. Now, sometimes that's necessary, but either way, it means more currency creation. So, the Fed it began as this sort of sober hallway monitor, but now they're often the responsible party that needs monitoring. But no. No one is going to stand up and do it because no one in power wants austerity under their watch because that is extremely unpopular. The third reason for permanent inflation is the Bretton Woods Agreement. You've probably heard of this, but let me summarize what it briefly means. Okay, Bretton Woods was the 1944 deal that basically created the post-World War II global monetary system? It made the U.S. dollar the world's reserve currency. If you remember anything from Bretton Woods, just remember that it did that. It made the U.S. dollar the world's reserve currency, and the dollar was pegged to gold at $35 per ounce.   Keith Weinhold  13:29   And finally, the fifth reason for never-ending inflation post World War II is Keynesian economics. I mean, you probably at least heard the term before. It's been thrown around here from time to time. Named after John Maynard Keynes, K E Y N E S. And before I go on, I invested in real estate for a long time before I learned all this stuff. Probably close to a decade of investing first. So I taught myself this material, Keynesian economics. That's the belief that demand is what drives economic output and employment. So, if you only remember one thing about Keynesian economics, it's that you need demand, and it stokes demand. It says demand drives everything, and what I mean by that is the spending, spending from households, corporations, and government. So, in plain English, when private demand weakens, the government should step in and spend. That's what Keynesian economics says. Well, that means deficits, borrowing, stimulus, support, programs, relief, rescue packages, emergency measures, and see what happens is that temporary measures somehow become permanent measures wearing a fake mustache. Remember, even Nixon said removal from the gold standard is temporary. Well, that was now 50. 55 years ago, in theory, the government runs deficits in bad times and then tightens up in good times. But that doesn't really happen because, in practice, government often runs deficits in bad times and good times, war times, peace times, election years, non-election years, leap years, all the time running deficits, spending more than we take in, and when deficits become normal, well, then currency creation has got to follow. That's the consequence. Well, these five forces that I told you about for never-ending inflation, the reasons that I just shared with you-they are now structurally embedded. They are not going away.   Keith Weinhold  19:03   I mean, there is even political resistance to deflation in this system. Investors benefit the most when they own one thing: real assets tied to long-term debt. You probably knew that I was going to say that because if the dollar is designed to slowly melt. You don't want to be the one holding the ice cube. You want to own the freezer. That's the control that you have. The first half of the year recently ended. It's time for our asset class rundown. From the midpoint of last year to the midpoint of this year, single-family home values are up only about one and a half percent. That's the average of Case-Shiller and FHFA. Apartment building values are down 1% in the past year. When it comes to rents per Zillow, single-family home rents are up 2.8% in the past year to an all-time record of almost 20-$300 Apartment rents are up just. 1.3% nationally. Sunbelt Apartments were the weak spot. Apartments.com said the South was down seven tenths of 1% year over year, and the mountain region down one and a half percent. With San Antonio, Denver, Austin, and Phoenix among the weaker markets, that's due to oversupply in those areas. 30-year mortgage rates down from 6.8 to 6.6% The S S&P 500 up 21 percent on AI optimism, despite a war in Iran. Though down in past months for the year, gold is still up 21 percent, silver soared 63 percent, Bitcoin down 45 percent. I mean, speculative digital assets have really gotten a cold shoulder. Oil up 4% although it went on a wild ride, and CPI inflation reheated to 4.2% That's our asset class rundown.   Speaker 2  22:59   This is our rich dad poor dad author Robert Kiyosaki. Listen to Get Rich Education with Keith Weinhold. Don't quit your daydream.   Keith Weinhold  23:17   Welcome back to Get Rich Education. I'm your host Keith Weinhold. I want you to listen to something along with me, and then I'll come back to comment. This is from the parallel truth. It's called the last generation to live normally, and it's less than two minutes in length.   Speaker 2  23:32   We have to talk about something that sounds dramatic, but it is becoming true. Your parents may have been the last generation to live a normal life-not an easy life, not a perfect life, but a life where the basic deal still made sense. You could get a stable job, you could buy a house, you could raise children, you could save some money, you could retire one day. And even if life was hard, most people still believed that if they worked honestly, their future would slowly get better. But look at what happened to your generation. You work more, but own less. You study more, but feel less secure. You have more technology than any generation in history, but less peace, less time, and less confidence about the future. Your parents were told, "Work hard, and you will build a life. But you are being told that, "Work hard, and maybe you can afford rent. And the most disturbing part is that this did not happen overnight. It happened slowly. First, housing became an investment instead of a basic need. Then, education became a debt trap. Then, healthcare became too expensive. Then, stable jobs disappeared. Then, everything became a subscription: your house, your car, your software, your entertainment, even your future. Everything slowly became something you rent but never truly own. And while ordinary people were falling behind, the economy kept looking strong on paper. The stock market went up, billionaires got richer, companies made record profits. Politicians kept saying that everything was fine, but if everything is fine, why does an entire generation feel like it is drowning? The truth is, your parents did not live through normal history. They lived through a rare window where ordinary people. People were allowed to share in the wealth of the system, but that window is now closing. The old promise was simple: work hard, buy a home, raise a family, retire with dignity. The new promise is different: work forever, rent everything, delay children, carry debt, and call it freedom. So maybe young people are not lazy. Maybe they are just the first generation honest enough to admit that the old deal is dead. Your parents were not lucky because life was easy. They were lucky because they were the last ones who got the deal before it was taken away.   Keith Weinhold  25:27   Yeah, there it is-the last generation to live normally. That's really a fresh slant on the hollowing out of the middle class. The rules have changed. Inflation is entrenched. Now you know why. Back in 2020, the pandemic accelerated that effect, and yet it's just unbelievable to me that people think working hard and saving money is enough to get you the lifestyle that you desire. Now I am not against hard work, it's the fact that people think that that's all that it takes. Before we hit the permanent inflation era, it might have made sense for you to say, save your money, pay all cash for a cheap fixer-upper property, and work hard for years to fix it up yourself. Oh, and then you could own a modest home debt-free. Today, even if you could do that, why would you? Instead, you can just prudently finance your way through life. You could have instead borrowed for two or three already renovated properties and let debt, inflation, and perhaps even tenants do the work for you. Above all, do the right thing before you do things right. That's what I like to say. Well, the way you get wealthy is by owning a lot of assets, not by grinding in the salt mines to pay off your debt. Those that are debt free are often asset poor. The biggest financial expense that you will ever have in your life. Do you know what it is? It is not taxes or interest charges. It's not even inflation or housing or healthcare or having children, most people have never heard of it. You probably have, but most people have never heard of this biggest financial expense you'll ever have, and they certainly don't know how to avoid it.   Keith Weinhold  27:34   Say that you're 35 years old and you put 100k under a mattress for 30 years until you're 60- years old. Instead, if that would have been invested at a 12% annual return, do you know how much that would have grown to? That would have grown to $2.996 million All right, basically 3 million bucks, a 30x increase. Therefore, it would be a 2.9 million dollar mistake to save money, and what this means is that the biggest expense you'll ever pay in your life is called opportunity cost. Yeah, opportunity cost is life's biggest expense. It's the return that was foregone when you chose one option over another. So opportunity cost is not what you spend; it's what your money could have become had you put it somewhere more productive. All right, now that was a pretty extreme example of 100k under a mattress. As a listener to this show, you are probably more savvy than a person that would save big lumps of money for close to zero return. Let me give you a better example of how when you pay all cash for something, you've usually just made your future self poorer. A friend of mine heard the episode last year where I talked about buying a new car for myself, a BMW X3 SUV. As it is, you probably remember that episode. Though I could have paid all cash for the car, I put the minimum down payment in there and then financed as much as I could because of a favorable 4% interest rate that I got on a car loan. Well, my friend Jesse heard that episode. This influenced him. So what he did is he bought a Subaru for his wife. Although he had planned to pay all cash and could have paid all cash for the car, Jesse got financing, and he did better than me. He got just a 1% interest rate somehow. Wow! It was actually nine tenths of 1% but let's just call it 1% What a deal! Instead of paying all cash for the car, he held on to that chunk of money. Instead of tying it up in a depreciating asset, he is financing it all. Now I don't. How much the Subaru costs, but let's just say it was 50k to keep the numbers simple. Well, look, if Jesse feels like he can get a 10% return over time by investing his money instead of sinking it into a car, how much does he profit by borrowing? Of course, he has the advantage of keeping his funds more liquid as well, but how much does he actually profit from this arrangement?   Keith Weinhold  30:24   Well, the math is so easy that you can even visualize it in an audio format here. Now it depends on the loan term, but the simple spread is a 10% investment return minus a 1% car loan cost. That is a 9% positive spread on 50k. That's roughly $4,500 per year in benefit. That's before any taxes, risk, or fees. $4,500 a year just for doing some loan paperwork. Like if you wonder whether the loan paperwork is worth it or not, that's what we're talking about here, and that's 375 bucks a month. So if you're wondering if it's even worth it taking the time to get a car loan when you could pay all cash, it probably is. All right, now that's the upside. What about the risk that's associated with taking a loan instead of paying all cash, well, the caveat here is that the 1% loan is guaranteed, but the 10% return is probably not, and that risk gap does matter. If you're financially fragile and you can't make the payment with another pot of money, well, then you risk default. That is over leverage risk. That's the worst case scenario. All right, what's the flip side? The flip side is that you could earn a return even better than 10% As we know, with real estate pays five ways on investment property. If you earn a 20% return, now you're making $9,500 a year on the spread, not $4,500, but a 10% return. That is the base case. So again, by paying all cash instead of getting the loan, your future self would be poorer by $4,500 a year. And now, my friend Jesse, that learned this from me, he's actually a CFA, a chartered financial analyst, a sophisticated money guy. But he had simply been overlooking this. And said another way, what you're doing here is that over time, your investment is paying you more than your interest is costing you, and in my life, I have been doing exactly this sort of thing all over the place for decades. An interesting thing that I hear about this, although it makes me scratch my head, I've heard a few people say this. It's just like, oh well, I don't want to have to deal with a car payment? I just rather be done with it and move on. What is there to deal with? Just set up auto pay with preserving funds for say a 10% return. You're then going to see more dollars flowing into your account than you will out of it. I mean that part can just be automated.   Keith Weinhold  33:19   My life and finances are set up this way. In fact, when I get a loan for a rental property, I have had mortgage loan officers that are looking at my finances. They tell me that I have more stuff flowing into and out of my checking account than they've ever seen anyone have. I'm I'm financing and arbitraging my way through life passively. This is thanks in part to inflation. I am not paying very much at all in that biggest financial expense that we all have in our lives-not taxes or children or housing, but opportunity cost. I am avoiding paying that. This is the world that we live in today, a lot of times debt reduction is horrible advice. Debt free that can keep people from falling over a cliff, but it stalls any wealth creation. Now the debts that usually make the most sense to pay down they're the ones with high interest, variable rates, no tax benefit, and no productive asset attached. And here is the priority order that I use for paying down debt or paying off debt. First, it is credit cards. Pay down these first almost every time. I mean, a 20% or even 30% credit card rate. This is like financial quicksand. You don't need a sophisticated investment thesis when you can get a guaranteed 20-4% quote-unquote return by eliminating this debt. The next place I would pay down are payday loans, personal. Loans and consumer finance debt. I mean, these are usually bad debts because they're at a high rate, have a short amortization, and they're usually tied to consumption instead of an income-producing asset. Pay these aggressively too, and then next in priority is paying variable rate debt that could reset higher. This isn't quite as important to address.   Keith Weinhold  35:24   We're talking about things like HELOCs, adjustable rate loans, margin debt, and some business lines of credit. Some of those can become dangerous when rates rise, even if the rate's tolerable today. The uncertainty can be a bit of a problem. Now, when it comes to should you pay down student loans, consider that. low fixed-rate student loans that might not be urgent. It sure wasn't for me. High-rate private student loans that could be different. That could get more of your attention. You also got to weigh things like tax benefits. Look out for forgiveness programs when it comes to student loans, those haven't been quite as available lately under this administration. Also, look at employer repayment benefits before you rush to pay down student loans, and then really the last one: low fixed-rate mortgage debt. Pay that last if you ever do. In fact, it is quite possible that I will always keep this debt type around that low fixed rate mortgage debt. So really, my rule of thumb here is to kill toxic debt. Be careful with unstable debt, and don't rush to pay off cheap fixed productive debt if you ever pay it off at all. You and I covered a lot of ground today, starting with 75 cent gasoline in California, all the way to the biggest expense you'll ever pay throughout your life, being something that most people have never heard of: opportunity cost. Coming up on the show here, a lot of good episodes, including a great guest and I are going to discuss a new way to invest in residential real estate that we haven't discussed before, and it will massively boost your cash flow. If you found today's show valuable, whether it was the history of why we have permanent inflation or the idea of passively financing your way to wealth, rather than only working harder. I would be grateful if you share this episode with a friend. Just tap the share button in Spotify, Apple Podcasts, or wherever you listen, and send it to someone who would benefit from hearing it. Or take a screenshot of this episode and post it on social media. It helps more people find the show, and it gives you and your friends something smart to talk about with each other. Until next week, I'm your host Keith Weinhold. Don't quit your daydream.   Speaker 1  37:53   Nothing on this show should be considered specific, personal, or professional advice. Please consult an appropriate tax, legal, real estate, financial, or business professional for individualized advice. Opinions of guests are their own. Information is not guaranteed. All investment strategies have the potential for profit or loss. The host is operating on behalf of Get Rich Education LLC exclusively.    Keith Weinhold  38:21   The preceding program was brought to you by your home for wealth building at getricheducation.com.  

This is not a growth hack. It is not going to go viral on a Twitter thread. But I genuinely believe it is one of the highest leverage things you can do for your brand right now, and most founders never do it properly because it is not exciting. Here is what a mentor told me years ago that I keep coming back to: a dollar saved is a dollar earned. And depending on your margins, that dollar saved is probably worth $1.30 or $1.40 on the bottom line. In this episode, I walk you through a full line by line expense audit covering every major cost area in a typical e-commerce business, the same process we have run at Foundr that has saved us tens of thousands of dollars a month. Here's what you'll take away: Why the average growing Shopify store is paying for 15 to 30 apps but actively using only eight of them, and how to fix that fast How to negotiate your SaaS tools, 3PL rates, merchant fees, and supplier costs in ways most founders never think to try Why agency retainers are one of the most expensive line items you can cut, and what to build in-house instead How to use AI and Claude Code to replace tools and creative spend that is quietly draining your budget every month The Meta ads Net 30 arrangement that can make a significant difference to your cash flow if you are spending at scale Why businesses waste an average of 26% of their marketing budget on campaigns that are not performing, and where to redirect it If your margins are tighter than they should be or you have not done a proper audit in the last six months, this episode will show you exactly where to look and what to do about it. If you're loving this solo series, I'd love to hear your feedback. Email me directly at nathan@foundr.com — I read every reply. Hope you enjoy it. WANT TO GROW YOUR BRAND WITH META ADS? Join the Foundr Operators Waitlist → ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://foundr.com/operators⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ HOW WE CAN HELP YOU SCALE YOUR BUSINESS FASTER Learn directly from 7, 8 & 9-figure founders inside Foundr+ Start your $1 trial → ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.foundr.com/startdollartrial⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ PREFER A CUSTOM ROADMAP AND 1-ON-1 COACHING? → Starting from scratch? Apply here → ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://foundr.com/pages/coaching-start-application⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ → Already have a store? Apply here → ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://foundr.com/pages/coaching-growth-application⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ CONNECT WITH NATHAN CHAN Instagram → ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.instagram.com/nathanchan⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ LinkedIn → ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.linkedin.com/in/nathanhchan/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ FOLLOW FOUNDR FOR MORE BUSINESS GROWTH STRATEGIES YouTube → ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://bit.ly/2uyvzdt⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Website → ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.foundr.com⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Instagram → ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.instagram.com/foundr/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Facebook → ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.facebook.com/foundr⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Twitter → ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.twitter.com/foundr⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ LinkedIn → ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.linkedin.com/company/foundr/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Podcast → ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.foundr.com/podcast⁠

Not all UV technologies are created equal, and for supply chain teams, knowing the difference leads to better purchasing decisions. In this episode of "Evidence Over Expense," Dr. Sarah Simmons, DrPH, CIC, FAPIC, and Juan Gonzalez from Xenex help cut through the confusion and explain what healthcare organizations should really be looking for before investing in UV technology. From FDA authorization and product safety to service, support, and long-term usability, this conversation gives supply chain teams a clear path to smarter decision-making. If your team is evaluating UV technology, this episode will help you ask the right questions, avoid the wrong assumptions, and look beyond the price tag to what truly drives value. BONUS CONTENT: Be sure to download the free Supply Chain UV Checklist Tool to help your team ask the right questions and make more informed purchasing decisions. Click here to download: https://9231499.fs1.hubspotusercontent-na1.net/hubfs/9231499/Power%20Supply/Podcast/Xenex%20Bonus%20Content%20-%20Episode2.pdf Once you complete the interview, jump on over to the link below to take a short quiz and download your CEC certificate for 0.5 CECs! – https://www.flexiquiz.com/SC/N/ps-xenex-ep2 A special thanks to our sponsor, Xenex, for making this series possible. #PowerSupply #Xenex #HealthcareSupplyChain #InfectionPrevention #ValueAnalysis #EvidenceOverExpense #UVTechnology #Podcast

Summary: The Book of Ephesians has been called the Swiss Alps of the New Testament as we have scaled great heights of God's love and calling upon our lives. And as the Epistle began, so does it end, with “Grace”. God's story, and your story, all begins and ends with Grace…. God's-Riches-at-Christ's-Expense. Living in Grace and giving Grace to one another……. We will live a life that takes us into the heavenlies, where Christ is seated! Join us as we end where we began….. In Grace.

Welcome to the Knives Templars Podcast—the show where blade enthusiasts, makers, and collectors unite! Each episode dives deep into the art and science of knife making, the stories behind legendary blades, and the vibrant community that keeps the edge sharp in the world of cutlery. Whether you're a seasoned smith, a passionate collector, or just discovering the allure of handmade knives, this podcast is your go-to resource for inspiration, education, and connection.A huge thank you to our incredible sponsors who make this show possible:·       EvenHeat Kilns – Precision heat treating for serious makers·       TR-Maker – Innovative tools for next-level knife crafting·       Pop's Knife Supplies – Your one-stop shop for premium materials·       Brodbeck Ironworks – Grinders and gear built for makers·       NJ Steel Baron – Steel that shapes legends·       Phoenix Abrasives – Abrasives that rise to the challenge·       KH Daily Knives – Blades and tools forged with passion·       Clark Iron Forge – Blacksmithing tools that strike true·       The Knifemakers' Guild – Craftsmanship, community, and traditionYou can catch the Knives Templars Podcast on all major platforms—Spotify, Apple Podcasts, Amazon, iHeart, Castbox, and wherever you get your audio fix. Be sure to subscribe, leave a review, and share with your fellow makers. Also see us on Facebook at the Knives Templars!https://knivestemplars.comBe Blessed

Quint and Allie break down the cost of investments and what to watch out for with each investment you buy. Hosted by Simplecast, an AdsWizz company. See https://pcm.adswizz.com for information about our collection and use of personal data for advertising.

When we talk about infection prevention, the conversation usually starts with clinical outcomes. But what if one of the most important decisions happens long before a patient enters the room? In Episode 1 of our brand-new podcast series "Evidence Over Expense," Dr. Sarah Simmons, DrPH, CIC, FAPIC, from Xenex and Steve Egbert from XENDELLA join us to explore how healthcare teams can take a smarter, evidence-based approach to evaluating UV technology. From manual cleaning limitations to FDA authorization and the real cost of pathogen transmission, this conversation helps supply chain teams look beyond the price tag and focus on what really matters before making the investment. Because behind every cleaner room, stronger workflow, and smarter technology decision is a supply chain choice with real impact. Once you complete the interview, jump on over to the link below to take a short quiz and download your CEC certificate for 0.5 CECs! – https://www.flexiquiz.com/SC/N/ps-xenex-ep1 A special thanks to our sponsor, Xenex, for making this series possible. #PowerSupply #Xenex #HealthcareSupplyChain #InfectionPrevention #ValueAnalysis #EvidenceOverExpense #UVTechnology #Podcast

Justin Spillers talks about how to leverage NOI, the ultimate driver of property value by focusing equally on raising rents and slashing costs. Justin breaks down the precise tactics you can use, from heavy value-add renovations and innovative revenue streams like pet rents and Wi-Fi surcharges, to negotiating bulk vendor deals and minimizing repair expenses. He shares the exact math behind ROI-driven upgrades, showing how a $15,000 renovation can generate a $36,000 annual increase in revenue, boosting your property's valuation exponentially at refinance. Justin Spillers Partner & Manager of Real Estate Alpha Based in: Minster, Ohio Where to find them: https://www.linkedin.com/in/justinspillers/ realestatealpha.io/ Book your free demo today at bill.com/bestever and get a $100 Amazon gift card. Visit https://malabarhillcapital.com/ for more info. Podcast production done by⁠ ⁠Outlier Audio Learn more about your ad choices. Visit megaphone.fm/adchoices

Operations are the hidden variable behind every multifamily return — and most investors never ask about them. This week, Senior Vice President of Gray Residential Katrina Greene joins Griffin and Blake to pull back the curtain on what actually happens between acquisition and exit.• What LPs should be asking about property management — but usually don't• How operational decisions protect or erode investor capital during the hold• Expense management strategies in a persistently inflationary environment• The maintenance blind spot costing multifamily owners more than they realize• How sticky residents are built — and why retention is driving revenue growth right now• First look at Gray Capital's two new acquisitions: Fairmont in Columbus, OH and The Century in West Lafayette, INWhether you're an LP evaluating sponsors, an industry professional, or a property management adjacent, this episode has something for you.

Yuval Refua is the Chief Product Officer at Navan, the global travel and expense platform he joined seven years ago when it was still just a travel booking service. Since then, he has built out its payments and expense products from the ground up, turning the company policy that used to live in a PDF into code that runs on the card itself. This conversation matters because T&E is one of the most universally disliked workflows in business, and Navan is rethinking it from scratch just as AI and agentic commerce start to reshape how companies spend.What We CoveredFalling in love with credit cards at American ExpressWhy Navan started as a travel-only booking serviceThe reconciliation pain that led to launching a cardCoding company policy directly onto the cardReal-time approval the moment you swipeWhy travel-first beats procurement-firstContext as the key to managing distributed spendGoing global with VAT, GST, per diems and mileageThe e-invoicing wave hitting more countriesThe GTA model for revealing complexity graduallyThe Expense Admin Companion and recommended actionsFrom single approvals to bulk to full automationThe Visa partnership and the Connect productWaymo for travelers, Formula One for financeKey TakeawaysThe expense report exists to answer a question that company policy already settled. Coding that policy onto the card removes the work instead of automating it.Starting from travel gives Navan context (where the employee is, why they are there, who they are visiting) that procurement-first tools lack, which makes per-employee limits far smarter.Going global is less about features and more about mastering country-by-country tax, e-invoicing, per diem and mileage rules.The path to full automation runs through trust. Navan moves finance teams from a single recommended action, to bulk approvals, to hands-off automation, which is also how it intends to handle agentic spend.About Yuval RefuaYuval Refua is Chief Product Officer at Navan. He started two companies of his own early in his career before moving into fintech and product management at Thomson Reuters, then American Express, where he developed a deep love for credit cards and the rails behind them. He joined Navan around seven years ago and has built out its payments and expense products from the ground up.Connect with Fintech One-on-One:Tweet me @PeterRentonConnect with me on LinkedInFind previous Fintech One-on-One episodes

Tony Castronovo is a Simple CFO fractional CFO who has worked with nearly 50 clients across real estate investing and small business ownership. In this second appearance on the show, Tony joins host Christina Gutierrez to walk through a string of five-star client reviews and unpack the real stories behind them — the financial messes, the predatory debt, the overleveraged portfolios, and the moments when a third-party lens changed everything for a business owner.This episode is a case study deep dive. From a three-pronged real estate and hard money operation that needed entity restructuring to a fiber construction company bleeding $7,000 a week to MCA lenders to a multifamily investor with a highly leveraged portfolio that needed property-by-property triage, Tony breaks down exactly how Simple CFO approaches each situation, why the CFO relationship only works when clients show up ready to collaborate, and what separates a bookkeeper from a financial partner who actually moves your business forward.Timeline Highlights[0:23] Tony Castronovo returns for his second episode — Christina introduces the format: unpacking real client reviews and the stories behind them[2:13] Tony's philosophy on celebrating wins, big and small, and why good news is worth sharing[3:34] Client one: Mike and Bill — a three-pronged business (traditional rentals, storage facilities, and hard money lending) all running through one entity when they arrived[5:26] The core pain when they came in: no cash flow clarity, no visibility into which business was making money and why[6:11] How Simple CFO handled pass-through revenue differently across three business models, and why the hard money business requires a completely different financial lens than storage or rentals[7:35] Entity restructuring with a CPA partner: separating the businesses for tax advantages, asset protection, and anonymity[8:01] Getting strategic once the basics are in place: the infinite banking play Tony introduced to help Mike and Bill finance storage unit purchases from their own policy instead of a lender[9:35] Why Simple CFO always starts with an expense analysis — and why every cut has to have an action attached to it, not just a number on a spreadsheet[11:11] The gym analogy: why Profit First implementation feels uncomfortable at first, gets routine, and then needs to be deliberately scaled up — just like adding weight once the reps get easy[13:52] Client two: Harley and Alex — came in effectively in crisis mode, overwhelmed by high-interest debt from predatory MCA lenders[15:30] The fiber construction business model: laying lines for carriers, owning and leasing equipment, and multiple revenue streams — plus multiple ways to spend money[17:07] How Simple CFO brought in a specialist with templated MCA negotiation scripts, saving Harley and Alex $7,000 per week in interest — roughly $30,000 a month[18:43] The snowball effect in reverse: freeing up capital, auditing the equipment inventory for bad debt, and building a path toward traditional financing[21:55] Deep dive on Alex's wife Claudia's equipment leasing business: reverse engineering the margins to find the keep number and identify exactly where gross profit was leaking[24:33] The Simple CFO network advantage: how Tony made a connection between a traditional flipper transitioning into cloudy title deals and an existing client already operating in that space[27:14] Business credit profiles: why most owners know their personal credit score but have no idea what their business credit profile looks like — and why it matters for accessing cheaper debt[28:49] Client three: Brett Long — London Living, a multifamily operator with a highly leveraged portfolio who came in recognizing that hope is not a strategy[30:52] Going property by property: analyzing gross potential rent, expense base, NOI, and debt service to identify dogs that need to be pruned from the portfolio[34:25] A live example from a flipping client the day before: stacking properties side by side to find the gross margin spread, identify holding cost problems, and fix the underwriting going forward[37:01] Why bookkeeping is the foundation of all of this — and the key difference between a bookkeeper recording transactions and a CFO using those records to make strategic decisions[39:27] Tony on what drives him: taking the financial stress off business owners so they can focus on the business they actually wanted to build[41:13] Christina's closing pitch: what to do if you hear these stories and recognize yourself in any of themKey TakeawaysClarity before implementation. Most clients arrive feeling like they're making money but not seeing it in their bank accounts. Simple CFO always starts with financial clarity — knowing the numbers — before designing any Profit First structure. You can't set allocations if you don't know what you're actually spending.Expense analysis is not academic. Every line item reduction needs a real action attached to it, and a CFO's job is to hold clients accountable to those actions between meetings. The results come from follow-through, not from a clean spreadsheet.A CFO relationship is a collaboration, not a fix-it service. Clients who come in wanting to be fixed don't get the same results as clients who come in ready to take action. The best outcomes happen when both sides hold each other accountable and trust flows in both directions.When predatory debt is bleeding the business, fix that first. Implementing Profit First while MCA lenders are taking weekly draws is adding structure to a system that can't sustain it. Tony's sequencing — stop the bleed, then build the foundation — is a deliberate order of operations, not a delay.The biggest portfolio is not the best portfolio. The most profitable portfolio is. Tony walks multifamily clients through a property-by-property NOI and debt service analysis to find underperformers that need to be pruned. Holding a cash-sucking asset because you're emotionally attached to it is a decision a third-party lens can fix.Your business credit profile matters more than you think. Most owners know their personal FICO score and nothing about their business credit profile. Improving that profile is what unlocks access to traditional, cheaper financing — and it often only takes a specialist and a plan to get started.Hope is not a strategy, and data is. Whether it's running a postmortem on every flip to analyze gross margins by property or building an underwriting template that tells you the max acquisition price before you ever talk to a seller, the CFO role is to replace optimism with actual numbers.Links & ResourcesSimple CFO (discovery call and reviews) — https://www.simplecfo.comProfit First for Real Estate Investors (free copy) — https://www.profitrei.comClosingIf any of the stories in this episode sounded familiar — the single-entity tangle, the MCA spiral, the overleveraged portfolio, the bank account that doesn't match what you think you're making — that's exactly who Simple CFO was built for. Tony and the rest of the CFO team run the same process, the same roadmap, and the same accountability system with every client. To read the reviews yourself or book a free financial discovery call, visit profitrei.com.

Should you expense a rental property cost immediately or capitalize and depreciate it over time? It's one of the most misunderstood areas of real estate investing and getting it wrong can cost you thousands in missed deductions or IRS headaches. In this episode, Thomas Castelli and Nate Sosa break down the decision framework every real estate investor needs to understand when dealing with repairs, renovations, improvements, appliances, HVAC systems, roofs, and other property expenses. You'll learn: - When an expense can be deducted immediately - How the De Minimis Safe Harbor works - The difference between repairs and capital improvements - When the BAR Test applies (Betterment, Adaptation, Restoration) - How cost segregation impacts your deductions - Bonus depreciation vs. Section 179 and when each makes sense - Common tax myths that trip up landlords and short-term rental owners Request a consultation from Hall CPA at go.therealestatecpa.com/3KSEev6 Get the FREE Ultimate STR Tax Strategy Bundle: go.therealestatecpa.com/strbundle Register for the FREE Investing Debate: go.therealestatecpa.com/debate Submit your question for Tom & Nathan: go.therealestatecpa.com/question The Tax Smart Real Estate Investors podcast is for general information purposes only and is not intended to provide, and should not be relied on for, tax, legal, or accounting advice. Information on the podcast may not constitute the most up-to-date legal or other information. No reader, user, or listener of this podcast should act or refrain from acting on the basis of information on this podcast without first seeking legal and tax advice from counsel in the relevant jurisdiction. Only your individual attorney and tax advisor can provide assurances that the information contained herein – and your interpretation of it – is applicable or appropriate to your particular situation. Use of, and access to, this podcast or any of the links or resources contained or mentioned within the podcast show and show notes do not create a relationship between the reader, user, or listener and podcast hosts, contributors, or guests. Any mention of third-party vendors, products, or services does not constitute an endorsement or recommendation. You should conduct your own due diligence before engaging with any vendor.

Execution drift rarely shows up as one big mistake. I've found that it starts with small deviations that seem harmless in the moment but eventually turn into bigger problems. For leaders, operators, and business owners, the real cost is not frustration or disappointment. It's the money, opportunities, and performance that slowly disappear when standards are not consistently enforced. In this episode, I break down the early warning signs of execution drift and how to catch them before they become expensive problems.  Show Notes: [02:32]#1 Drift compounds into hidden financial loss. [09:16]#2 Drift slows decision cycles and kills leverage. [12:22]#3 Drift erodes trust internally and externally. [16:17] Recap Next Steps: --- Execution is not a talent.   It is a standard. If your results don't match your ability, something in your approach is out of alignment. Most people do not have a motivation problem.   They have a consistency problem. Power Presence is the system for operating with greater discipline, clarity, structure, and execution under pressure. Learn more: → http://www.PowerPresenceProtocol.com  — This show is the public record of standards. All episodes and the complete archive: → http://WorkOnYourGamePodcast.com 

0:30 - Teen takeovers in Chicago 16:37 - Iran 44:54 - Remembering Tom Dreesen: Dan’s interview with Tom from 11/7/25 01:18:05 - Professor at George Mason University Scalia Law School and senior fellow at the Heritage Foundation, Eugene Kontorovich, weighs in on the Memorandum of Understanding, saying “It strengthens Iran, there is no other way to put it.” Follow Professor Kontorovich on X @EVKontorovich 01:36:35 - University of Chicago law professor emeritus Richard Epstein discusses his legal battles over the Obama Presidential Center, saying, “If you’re 100% right in a case against the government, you have a 50% chance of winning.” Check out Richard’s newest book The Myth of Birthright Citizenship 01:53:54 - Manhattan Institute researcher Neetu Arnold discusses grade inflation and why schools may need new ways to separate exceptional students from the pack. 02:07:32 - Hussain Abdul-Hussain, research fellow at the Foundation for the Defense of Democracies, on the Iran peace deal and Trading Away Lebanon: Washington’s Bargains at Beirut’s Expense. Hussain is also the author of The Arab Case for IsraelSee omnystudio.com/listener for privacy information.

Send us Fan MailHow do you scale a Home Services Business past the $10M mark without losing your Company Culture or selling out to Private Equity? In this episode of Let's Vent, we sit down with the Owners of Go Green Plumbing, Heating & Air, Alicia Green and Pete Green to break down the exact operational tips, strategies and ideas they implemented to build an independent trade powerhouse. Connect with out Guests: Go Green Plumbing: https://gogreenplumb.com/Alicia Green: https://www.linkedin.com/in/alicia-green-14bb3495/Pete Green: https://www.linkedin.com/in/pete-green-25496072/ Connect with our sponsor: https://freeagency.aiTime Stamps: 01:10 - Introducing Pete & Alicia Green from Go Green Plumbing02:18 - Pete Green's Transition from Programming to "Chief Technology Officer"03:45 - The Truth About Company Culture: There are Always Ups & Downs05:15 - What Happens When People Don't Fit the Mold?06:13 - Shifting from Professional to Lightheartedness in Tough Times07:45 - The Go Green Hiring Process: Do You Let Your Team Make the Decisions?09:20 - The "Princess Castle" Lego Challenge & Out Of Comfort Zone Testing13:16 - Quick to Hire, Slow to Fire: Should Be The Opposite Way Around? 14:15 - The ROI of Training16:55 - Joining Nextstar Network & Implementing Soft Skills Training17:58 - The Academy Structure: Weekly Breakdown of Trades & Certifications22:38 - 60% of Our Business Wouldn't Exist Without the Training Academy23:45 - The Myth of the Unicorn Employee25:03 - Balancing IQ and EQ: Why Technical Skills and Soft Skills Are 50/5027:50 - The Chaos of Early Training Programs vs. Today's Managed Structure29:15 - Building a Clear Pay Plan and Incentivized Levels32:14 - Advanced Lab Training: Partnering with Ultimate Tech Academy in Arkansas33:20 - The Tax Perspective: Are you Paying More? 34:00 - Facing Private Equity (PE) in the Trades38:12 - The Positive Side of PE: Injecting Business Logic and Real Value into the Trades42:50 - Growing Big with Zero Outside Capital45:15 - Why Cheap Prices Come at the Expense of Employees?47:50 - Built on Community assistance: The Go Green Community Promise Program49:35 - "Owned by Google": Venting About the Real Monopolies Dictated by the Industry53:48 - Where Does the Cash Flow? Canadian Agencies vs. Local Greensboro Wages57:25 - Understanding KPI Pressures and Employee Mass Exits58:35 - Why Technicians Stay for Culture and Run from Structure Changes01:01:40 - The Flaw in Flipping: Why Passing Hands Leads to Volume Loss01:05:43 - Processing the Reality of Multi-Billion Dollar Acquisitions in the Trades01:07:55 - The Challenge to Maintain Massive Service Value Over Time01:10:17 - Will AI Supplement or Completely Replace Modern Jobs?01:12:00 - How To Choose a Software That Actually Helps Reduce Workload?01:12:42 - Why the CTO's Workload Increases When Implementing AI?01:13:55 - The Sandbox Mindset: Starting From a Place of Natural Curiosity01:18:15 - Managing Scope Creep When Coding with Accelerated AI Speed01:21:38 - How Non-Technical Leaders Can Leverage Claude?01:23:59 - The "Twice a Day" Automation Rule: Building Your Operational Task List01:28:46 - The Importance of Context and Direct Communication to Create a Prompt correctly01:31:12 - Ostrich Mentality: Why Ignoring the Automation Wave Will Cost People Their Careers01:31:47 - Focus on Eliminating Time Rather Than Solving the World's Problems01:34:00 - How To Hold People Accountable Without Destroying The Company Culture ?01:34:48 - Facts Over Feelings: Gathering Documentation and Data Before Tough Conversations01:38:25 - Final Thoughts on How to Build a Scalable Organization

The FASB's disaggregation of income statement expenses (DISE) guidance requires public business entities to provide significantly more detail about key income statement expense captions beginning in 2027. This episode covers what the new disclosure requirements mean, why implementation may be more complex than expected, and how companies can start preparing their data, systems, processes, controls, and judgments now.For more on this topic read section 3.11 of PwC's Financial statement presentation guide and our publication, FASB issues new disaggregated expense disclosure requirements (DISE).Follow this podcast on your favorite podcast app and subscribe to our weekly newsletter to stay in the loop for the latest thought leadership on sustainability standards. About our guestsAngela Fergason is a partner in PwC's National Office. She is an experienced consultant on technical accounting and financial reporting matters, specializing in revenue recognition, employee compensation, and emerging issues impacting the technology industry. Angela is also PwC's standard setting leader, managing PwC's strategy for engaging in accounting standard setting activities.Gary Sardo is a partner in PwC's Deals practice who advises companies on accounting and financial reporting matters from acquisitions, divestitures, capital raises, and complex deals, particularly in the pharmaceutical and life sciences industry. In this role, Gary also supports companies navigate the implementation of new accounting standards and evolving financial reporting requirements. Recently, Gary completed a tour in PwC's National Office and a two-year fellowship at the Financial Accounting Standards Board.About our hostHeather Horn is the PwC National Office Sustainability and Thought Leader, responsible for developing our communications strategy and conveying firm positions on accounting, financial reporting, and sustainability matters. In addition, she is part of PwC's global sustainability leadership team, developing interpretive guidance and consulting with companies as they transition from voluntary to mandatory sustainability reporting. She is also the engaging host of PwC's accounting and reporting weekly podcast and quarterly webcast series.Transcripts available upon request for individuals who may need a disability-related accommodation. Please send requests to us_podcast@pwc.com. Did you enjoy this episode? Text us your thoughts and be sure to include the episode name.

Most people misunderstand whole life insurance because they look at it as a product instead of a system. In this Practical Wealth Study Group, Curtis May breaks down the Four Stages of Whole Life Insurance, also known inside the Money4Life Blueprint as the Private Reserve Strategy. This is not about chasing rates of return. This is about control, liquidity, certainty, and building a personal economy where your money keeps working inside your system instead of constantly leaving to banks, lenders, credit cards, and financial institutions. Curtis walks through the Money4Life Framework: Earn it. Bank it. Borrow it. Spend it. Repay it. Repeat. You'll learn how whole life insurance can function as a foundational asset, why premium should be viewed as a capital flow instead of an expense, and how families and business owners can begin using their policies to recapture debt, build liquidity, and eventually finance opportunities. This conversation covers: Why whole life insurance is not an investment account The economic value of certainty The crisis of financial control Why liquidity matters more than rate of return How to calculate your burn rate Why you must capitalize before you invest The difference between being a saver, wealth builder, business banker, and infinite banker How to stop giving interest away to strangers Why banking is a process of becoming, not a product you buy The goal is not just to own a policy. The goal is to become the banker. Visit PracticalWealth.net to take the Financial Freedom Assessment and learn more about the Money4Life Blueprint. 00:00 – Welcome to Practical Wealth Study Group 00:19 – The Four Stages of Whole Life and IBC 01:00 – Whole Life Is Not an Investment Account 01:45 – The Economic Value of Certainty 02:30 – Whole Life as a Foundational Asset 03:10 – The Money4Life Framework: Earn It, Bank It, Borrow It 04:20 – Why Banking Means Control of Capital 05:30 – The Crisis of Control 06:15 – Stop Giving Away the Banking Function  07:00 – The Maturity Matrix: Where Do You Stand? 08:00 – Stage 1: The Saver 09:20 – You Can't Invest Until You Capitalize 10:30 – Contract Wealth vs. Statement Wealth  11:45 – Stage 2: The Wealth Builder 12:45 – Premium Is Not an Expense 13:45 – Freedom From Debt to Others 14:40 – Your Burn Rate and Liquidity Number 15:50 – Debt-to-Capital: Bringing Debt In-House 17:00 – The Difference Between Chaos and Opportunity  18:00 – Stage 3: The Business Banker 19:00 – Money as Inventory  20:00 – Financing Opportunities Through Your System  21:00 – Stage 4: The Infinite Banker 22:00 – Closing the Financial Loop  23:00 – Banking Is Not a Product 23:30 – Immediate Action Plan

What happens when the drive that helped you build your success becomes the very thing that pushes you toward burnout? Many entrepreneurs pride themselves on powering through exhaustion, stress, and overwhelm. But eventually, the body sends signals that can no longer be ignored. In this episode, Rachel sits down with Dr. Anna Cabeca, a triple board-certified OB/GYN, bestselling author, hormone expert, and founder of a thriving wellness company. Dr. Anna shares how personal loss, infertility, burnout, and financial hardship led her to completely rethink health, performance, and what lasting success really looks like. In this episode, she breaks down how leaders can protect their energy, improve resilience, and build a healthier foundation for both life and business. When Success Comes at the Expense of Health Dr. Anna's entrepreneurial journey wasn't born from a business plan. It was born from necessity. After losing her mother, experiencing infertility and early menopause, navigating divorce, and eventually reaching severe burnout, she found herself forced to step away from the medical practice she had spent years building. The emotional and financial consequences were significant, but so was the lesson. Rather than accepting the limitations she was told to live with, Dr. Anna began exploring functional medicine, nutrition, and integrative approaches to health. Her own transformation became the catalyst for helping others do the same. What started as solutions for her patients eventually evolved into bestselling books, educational programs, and a seven-figure wellness brand built around solving real problems and sharing authentic stories. Why Energy is the Ultimate Asset One of the most powerful themes in this conversation is the connection between health and performance. Dr. Anna explains how chronic stress, elevated cortisol, inflammation, and insulin resistance can affect everything from decision-making and focus to mood, resilience, and long-term health. While many people focus solely on hormones, she argues that true wellbeing requires a broader approach. She also introduces the concept of increasing oxytocin, the hormone associated with connection, trust, and wellbeing, as a practical strategy for managing stress and supporting overall performance. Enjoy this episode with Dr. Anna Cabeca… Soundbytes 29:11–29:40 "You can't out-supplement a bad diet. So, it's these lifestyle pieces. So when I think about what are some of the things we do to improve our health? It's starting with — I don't like to talk about stress management — increasing oxytocin. That is the antidote to stress. So, gratitude practice. Positive terminology. Being kind to yourself. Being kind to others." 36:39–37:49 "I was over 240 pounds. I had terrible weight loss, and that was at 39 years old. Everyday I think, OK, I can walk towards health, or I can walk towards disease." Quotes "Your mess becomes your message." "It takes more than hormones to fix the hormones." "Workaholism is an addiction like anything else." "We have it within us to be empowered and to create solutions and serve others." "I want to be a safe place where people feel seen and heard and empowered and inspired that their life can be better." Links mentioned in this episode: From Our Guest Website: https://dranna.com 10-Day Breeze Through Menopause Program: https://dranna.com/tribe Connect with Dr. Anna Cabeca on LinkedIn: https://www.linkedin.com/company/drannacabeca Follow Dr. Anna Cabeca on Facebook: https://www.facebook.com/DrAnnaCabeca Follow Dr. Anna Cabeca on Instagram: https://www.instagram.com/thegirlfrienddoctor Connect with brandiD Find out how top leaders are increasing their authority, impact, and income online. Listen to our private podcast, The Professional Presence Podcast: https://thebrandid.com/professional-presence-podcast Ready to elevate your digital presence with a powerful brand or website? Contact us here: https://thebrandid.com/contact-form/

Another example of the problems with Annuities.Is it OK to pay higher fund expense ratios for higher returnsLaura Pausini concert in OrlandIoniq 9 and some EV newsStill time to go to CSI Con in New York

We all want retirement success. But how do we achieve it? What if the best method is to identify possible *failures* first, and then simply work backward to avoid those failures?  Looking for a financial planner?  → PlanWithJesse.com In this follow-up episode, Jesse completes his inversion-based framework for retirement planning by outlining the remaining risks that can derail long-term financial outcomes, shifting from market and inflation concerns to more personal, behavioral, and systemic threats. He begins with shock spending and long-term care risk, emphasizing the scale and unpredictability of end-of-life care costs and arguing that insurance alone is often insufficient, making realistic cash flow modeling and programs like Medicaid more practical planning tools. He then covers cognitive decline risk, highlighting how reduced decision-making capacity can lead to fraud, mismanagement, and financial error, and recommends safeguards such as legal protections, trusted contacts, and automated, simplified financial systems. Behavioral risk is framed as the danger of emotional decision-making, with mitigation strategies including automation, written investment policies, and reduced exposure to market volatility. Jesse then addresses assumptions risk, warning that small inaccuracies in assumptions about markets, inflation, taxes, or even one's future self can compound significantly in retirement projections, advocating for base rates and disciplined "what-if" analysis. He explores policy, legislation, and tax risk as an unavoidable layer of uncertainty around Social Security, taxation, and healthcare policy, suggesting retirees stress test outcomes without overreacting to speculation. Identity and purpose risk follows, underscoring that retirement success depends heavily on structure, meaning, and social connection, not just financial security. Finally, he introduces "deep risks"—deflation, confiscation, and devastation—arguing that while rare, these systemic threats reinforce the central conclusion that no portfolio design eliminates all risks, and effective retirement planning ultimately comes down to balancing trade-offs and building resilience. Key Takeaways: • Shock spending risk includes large, unexpected expenses that can destabilize retirement plans. • Long-term care is one of the most significant and unpredictable retirement costs. • Cognitive decline can lead to financial mistakes, fraud vulnerability, and poor decision-making. • Behavioral risk stems from emotional and irrational financial decisions. • Assumptions risk arises from unrealistic expectations about markets, inflation, or personal behavior. • Policy and tax risk includes uncertainty around Social Security, taxes, and healthcare programs. • Identity and purpose risk highlights the psychological challenges of retirement. • Deep risks (deflation, confiscation, devastation) are rare but potentially catastrophic. • No single strategy can eliminate all risks—retirement planning is about balancing trade-offs and building resilience. Key Timestamps: (01:42) – 8: Shock Spending & Long-Term Care Risk (08:04) – Saving for the Coming $500,000 Expense (09:15) – Changing Expenses as We Age (10:24) – Medicare & Medicaid (12:44) – 9: Cognitive Decline Risk (15:43) – Building Backup Systems & Backup People (18:30) – 10: Behavioral Risk (22:48) – 11: Assumptions Risk (About Yourself & the World) (25:18) – Assumptions About the Future World (31:50) – 12: Policy, Legislation, & Tax Risk (36:17) – 13: Identity & Purpose Risk (39:16) – 14: The Deep Risks Key Topics Discussed: The Best Interest, Jesse Cramer, Wealth Management Rochester NY, Financial Planning for Families, Fiduciary Financial Advisor, Comprehensive Financial Planning, Retirement Planning Advice, Tax-Efficient Investing, Risk Management for Investors, Generational Wealth Transfer Planning, Financial Strategies for High Earners, Personal Finance for Entrepreneurs, Behavioral Finance Insights, Asset Allocation Strategies, Advanced Estate Planning Techniques Mentions:https://bestinterest.blog/e108/ Stumbling on Happiness by Daniel Gilbert Thinking, Fast and Slow by Daniel Kahneman https://bestinterest.blog/the-crushing-cost-of-conservative-retirement-planning/ https://bestinterest.blog/e106/ If You Can: How Millennials Can Get Rich Slowly by William J. Bernstein The Intelligent Asset Allocator: How to Build Your Portfolio to Maximize Returns and Minimize Risk by William J. Bernstein A Splendid Exchange: How Trade Shaped the World by William J. Bernstein The Four Pillars of Investing, Second Edition: Lessons for Building a Winning Portfolio by William J. Bernstein Deep Risk: How History Informs Portfolio Design by William J. Bernstein More of The Best Interest: Check out the Best Interest Blog at https://bestinterest.blog/ Contact me at jesse@bestinterest.blog Need a financial planner?  → PlanWithJesse.com  The Best Interest Podcast is a personal podcast meant for education and entertainment. It should not be taken as financial advice, and is not prescriptive of your financial situation.

Join us as Pastor Craig takes us through Spent: Selling your Present at the Expense of your Future!

Michael Steele tackles the stark realities of a system that favors the wealthy while leaving the middle class and the poor to fend for themselves. With insider information flowing to the elite, decisions are made that enrich a select few at the expense of the many. Michael exposes the troubling dynamics at play, from stock market manipulation to the alarming disconnect between policy and the everyday struggles of American families. Tune in to understand how this rigged system is shaping the narrative as we head into the fall.Catch Michael Steele on The Weeknight Mondays - Fridays at 7pm EST on MSNBC: https://www.msnbc.com/weeknightFollow Michael on X: https://x.com/MichaelSteeleFollow Michael on Bluesky: https://bsky.app/profile/michaelsteele.bsky.socialFollow Michael on Instagram: https://www.instagram.com/chairman_steele/Follow Michael on Threads: https://www.threads.net/@chairman_steeleListen to The Michael Steele Podcast: https://podcasts.apple.com/us/podcast/the-michael-steele-podcast/id1412905534Watch The Michael Steele Podcast: https://www.youtube.com/playlist?list=PLJNKzTkCZE9uNqPiKYw5eU5YkS_mMsr6oIf you enjoyed this, share it with a friend!

What happens when one of the world's fastest-growing travel platforms decides the future of business travel will be built around AI from the ground up? In this episode of Tech Talks Daily, I sat down with Navan co-founder and CTO Ilan Twig to discuss how the company is reshaping travel, payments, and expense management through AI-native systems designed for the real world, not just polished demos. What immediately stood out during our conversation was Ilan's mix of technical obsession and relentless focus on user experience. This is someone who isolated himself for months to truly understand the mechanics of large language models before most companies had even worked out what ChatGPT meant for their business. That curiosity now powers Navan's AI strategy, where conversational interfaces are replacing what Ilan calls the old "forms and tables" model of software interaction. We explored how Navan's AI assistant, Ava, is already handling thousands of real-world travel support conversations every day, with customer satisfaction scores that rival those of human agents. During major disruption events like Storm Fern and the Heathrow airport fire, Ava scaled instantly, resolving huge volumes of customer requests without the delays and staffing nightmares that traditionally overwhelm travel providers. But this conversation goes much deeper than travel. Ilan shared his thoughts on why the software industry is moving toward conversational, context-aware interfaces, why most businesses still misunderstand what agentic AI actually means, and how Navan is building proprietary models trained on its own travel data to outperform larger, generic frontier models. We also discussed trust, hallucinations, AI supervision layers, and why companies must stop treating AI as a magic trick and start measuring it against hard business outcomes. There is also a fascinating human side to this episode. From building a company through market turbulence, investor skepticism, and geopolitical uncertainty, to challenging accepted thinking since his school days, Ilan's story reflects the mindset of someone who genuinely believes technology should solve real problems rather than create headlines. If you have been wondering where AI moves beyond hype and starts delivering measurable operational value, this conversation offers a rare look behind the curtain from someone building these systems at scale every single day. Useful Links Connect with Ilan Twig Learn more about Navan Check out blog posts by Navan Follow Navan on LinkedIn Visit our Sponsors Check out the Nordlayer Browser Learn more about Denodo Data Products  

The guys talk about the tragic passing of Kyle Busch, the Vegas talk heats up as the gang revisits the infamous forgotten, deleted podcast

The guys talk about the tragic passing of Kyle Busch, the Vegas talk heats up as the gang revisits the infamous forgotten, deleted podcastSee omnystudio.com/listener for privacy information.

Somehow, Mike didn't know how to pronounce "durag", and the guys let him hear about it.

Did you deduct actual expenses or use the standard mileage rate for your vehicle? More importantly, was that the best choice for your situation? Book a comprehensive tax and business consultation with Mark and Mats firm KKOS Lawyers to ensure your strategy works for you.Grab my eBook 30 Unique Strategies Every Business Owner Should Know! You don't want to miss this! Secure your tickets for the #1 Event For  Small Business Owners On Main Street America:  Main Street 360 Looking to connect with a rock star law firm? KKOS is only a click away! Are you ready to get certified in EVERY strategy I teach? Start your journey with a FREE 15-minute discovery call to explore the Main Street Tax Pro Certification. Check out our YOUTUBE Channel Here: https://www.youtube.com/markjkohlerCraving more content? Check out my Instagram!