Souk's Notes - Unofficial Summaries

2025-03-13 · open the original ↗

• Bob's "record" production gain under Trump is cherry-picked. In the clip, Bob cites 13.2 million barrels a day a year ago, 13.3 million at year-end and almost 13.6 million now, calling it more growth than all of last year; he counters that this compares a weekly figure with a year-ago point and an annual average, and output exceeded that level in December.
• Presidents do not set oil production, which has fallen since inauguration. He says production has fallen since Trump took office but that is not Trump's fault, just as 2024 output was not Joe Biden's doing; not one rig is working because Trump is president. He accuses Bob of deliberate misinformation to give Trump credit.

2025-03-13 · open the original ↗

• The Deep Learning Tuning Playbook aims to turn alchemy into chemistry. Godbole, its first author, says it pooled his Google Brain co-authors' decades of experience into a systematic, architecture-agnostic process: set up a good initial experiment, then decide each next experiment from the data collected, given a compute budget and constraints. He credits the co-authors as the real brains and calls it their "two cents", not the final answer.
• Releasing it as a forkable markdown file made it a community resource. Rather than a PDF on arXiv, the team published it on GitHub under a permissive license to invite forks and pull requests; community members crowdsourced translations the authors do not endorse. The host reports 28,000 GitHub stars at recording and says it answers the question his deep learning students always ask about why to add layers or neurons.

2025-03-12 · open the original ↗

• Energy Secretary Chris Wright has become a Trump mouthpiece. He says Wright, once a rational voice, told oil and gas executives at CERAWeek that shale could boost production even at $50 oil, a claim he says no one but Trump believes; he cites this as why he won't enter politics.
• The Alaska-versus-Iran restrictions claim misreads Biden's record. Answering a clip of Wright saying Alaska faced more restrictions than Iran, he argues Biden limited drilling only where no company wanted to drill, and approved the Willow project, one of Alaska's largest.
• Industry has no appetite for ANWR. He reads the Alaska line as setting the stage for ANWR drilling, and points out that oil companies placed not a single bid in the last two ANWR lease sales.

2025-03-12 · open the original ↗

• Countries don't trade energy; corporations do, for profit. Answering a viewer who asked why the US can't power the three states Canada supplies, he says cross-border energy flows reflect company choices rather than need, and adds that Mexico also sends electricity to some southern states and that Canadian oil and gas are a large part of the picture.
• The tariff fight is about Trump's need for attention. He argues the tariffs have nothing to do with the economy or fentanyl, noting Trump now trashes US–Canada–Mexico trade agreements that were his own.
• Republicans turned Canada into an enemy overnight. He says the party spent years naming China and Russia as the main threats, said nothing against Canada until Trump's inauguration, and that the US does not need this fight with its closest allies.

2025-03-12 · open the original ↗

• Open models win on control; closed models on performance and ease. Maarten Grootendorst, co-author with Jay Alammar of Hands-On Large Language Models and maintainer of BERTopic and KeyBERT, works at a cancer research institute that will not send patient data to any cloud provider. He adds other reasons to run models yourself: not knowing which version or business rules sit behind a proprietary API, content restrictions, outages, and fine-tuning on domain data no lab can see. Open tooling, often built by hobbyists, will "never ever be as perfect" as closed providers' interfaces. He names Llama, Mistral, Qwen and DeepSeek as strong open models.
• Small models will improve fast through distillation and serve agents. Large models generate training conversations that teach cheap small models, which already run on phones. In his agent picture, one large model steers many tiny ones that summarize or transcribe, so a model about 90% as accurate handles basic work without the cost of a large one.
• He predicts open source will always catch up with proprietary. He traces scaling from train-time compute to test-time compute ("more so-called thoughts"), and DeepSeek matched that and ranked in the top five, second or sometimes first depending on the evaluation. New architectures such as Mamba, BitNet and Google's Titans show promise, but he says only well-funded labs can test whether they scale, because a university will not "throw 50 million at the problem".
• Representation models still beat generative ones for many tasks. You embed, say, 10 million documents once and reuse the vectors for classification, clustering, topic modeling, deduplication and retrieval. Running those documents through ChatGPT would be slow and costly, and costly again every time the prompt changes. "We're not going to use GPT reasoning models to classify spam." He sees a recurring cycle of chasing the shiny model when bag of words, word2vec or a cheap BERT classifier would do.
• Communication is the most underrated skill in data work. Stakeholders care about usefulness, not model performance, and many data science teams must keep justifying their existence because they are advisory rather than a product, he argues. He suggests practicing by recording 30–60-second explanations on TikTok until viewers' questions shift from "I don't understand" to how the idea would be applied.

2025-03-12 · open the original ↗

• Oil's fall is sentiment, not fundamentals. He says oil has tracked the Dow down on recession fears, tariffs and uncertainty, while a chart of global balances he shows still has the market well under-supplied.
• The OPEC+ increase will net out as a supply cut. Traders forgot that Kazakhstan has overproduced its quota by about half a million barrels a day; once it complies, the 138,000 barrels a day OPEC+ increase leaves a net loss. US output is down about 150,000 barrels a day in recent months, and the EIA's STEO forecasts tightening through mid-year. Oil recovered a couple of percent to $67.67.
• Gasoline will stay above $3 despite cheaper crude. Even after oil fell more than $10, gas is $3.08 a gallon; he notes prices climb from the second or third week of March with the switch to summer blend, so only a complete oil collapse would bring a meaningful drop.
• Lost refining capacity has decoupled pump prices from crude. In Trump's first term, oil at $65 to $67 meant gas well below $3 a gallon; now reduced refining capacity keeps gasoline inventories too low and record refined-product exports add to it. He expects gas prices to rise from here.

2025-03-12 · open the original ↗

• Inflation-adjusted, oil is already at the old break-even price. In a clip, former Pioneer CEO Scott Sheffield says today's price equals $50 of 20 years ago, public companies' cash break-even including dividends is $50 to $55 a barrel, and US production fell in 2015-16 at true $50. He agrees: "$70 is the new $50."
• Producers stop drilling long before break-even. He argues a company holding, say, 10 million barrels of reserves will not sell at break-even when waiting could earn $20 to $30 a barrel more, and shareholders expect a real return, so break-even is the wrong threshold to discuss.
• Falling prices would collapse rig counts faster than ever. Rig counts are slipping but still within range; if prices keep falling he expects declines to accelerate quickly, because rigs now finish a well in two or three weeks rather than six months.
• Fundamentals will win over today's psychological selling. He says the entire drop is negative sentiment over Trump, tariffs and the economy, and that fundamentally oil should not be as low as $66 to $67 a barrel; traders are pricing a crumbling economy.

2025-03-12 · open the original ↗

• Republicans target the Education Department to deflect red-state failures. He says 24 of the bottom 25 states for education are red states, none of the top 10 are, and red states take more federal money than blue ones, yet Republicans blame the Department of Education and want schooling returned to the states.
• Trump can cripple the Department but not abolish it alone. He argues Trump can leave it largely dysfunctional for the rest of his term; blue states' funding would be little affected, but Title I and special education for children with disabilities would take the hit.
• He expects the red-blue education gap to widen sharply. Already massive, the disparity would grow if red states run education themselves, he argues, since in his view they do not prioritise funding it; parents there will need to step up far more.
• He predicts the Department survives because Congress protects its jobs. However loyal members of Congress are to Trump, he says, they are more loyal to their own seats and will not let it go.

2025-03-12 · open the original ↗

• This post includes paid content that is not available publicly. See the original source.

2025-03-12 · open the original ↗

• A deep learning system beat radiologists at predicting breast cancer from mammograms. Godbole, co-first author of the 2020 Nature paper "International evaluation of an AI system for breast cancer screening", says it predicted biopsy-confirmed cancers two or three years ahead, depending on the country's screening guidelines, better than expert radiologists. It was benchmarked on a large retrospective US and UK dataset of what he recalls as a hundred-and-some thousand patients, and beat every radiologist in a separate reader study.
• The bigger promise is combining AI and humans and widening access. Godbole says humans and AI have different strengths, so their synthesis can be powerful, though he is out of date on that literature. He highlights countries with many patients per radiologist, where AI could substantially improve access to potentially life-saving screening.

2025-03-11 · open the original ↗

• The "four-year low" gas claim is false on its own chart. He says the creator making it was standing beside a three-year chart, not a four-year one, and gas was cheaper for weeks only a couple of months earlier, which anyone can check at the pump.
• Attention-driven content has no incentive to be right. He contrasts creators who say anything for followers with his own aim of informing people, arguing even those who reject his politics acknowledge he is right on oil and gas, and asks how such claims would survive a live challenge.

2025-03-11 · open the original ↗

• Gas prices are exactly where they were on Election Day. Answering a supporter's clip celebrating Trump over gas prices, he cites the EIA average for all formulations at $3.19 a gallon on Election Day and $3.19 today; prices rose nearly 10 cents after inauguration and came back, which he likens to Democrats celebrating a similar retreat under Joe Biden.
• "Drill baby drill" is not happening; output has fallen. He says US oil production is down almost 200,000 barrels a day since Trump took office, and that even a genuine drilling push would take at least a year to show in the numbers.
• Only an economic crash brings gas down meaningfully. Repeating his long-standing view, he shows the Dow Jones and oil prices falling together as evidence that lower fuel prices are coming through a weakening economy rather than policy success.

2025-03-11 · open the original ↗

• The new U7 Pro XG and U7 Pro XGS are fanless redesigns. Both have a near-flat face and an all-metal back built to shed heat, so no fan and no noise, plus a 10 Gbps RJ45 port powered over PoE+. The new mounting plate is not a drop-in replacement, but it keeps the old screw-hole pattern, so no new holes are needed.
• The XGS is the higher-capacity model, with a dedicated scanning radio. The host gives the XG as 2x2 MU-MIMO on all three bands, about 1,500 sq ft and around 300 clients. The XGS is 4x4 MU-MIMO on 5 GHz (2x2 on 2.4 and 6 GHz), about 1,750 sq ft and over 500 clients, and scans continuously for interference to adjust channels in real time.
• The host calls them a smarter buy for futureproofing. He describes the XG as essentially a U7 Pro and the XGS as a U7 Pro Max, each with a 10-gig NIC and the new fanless form. He places them between the existing U7 and the higher-end E7 lines, and says the 10-gig NIC alone justifies the small extra cost. Prices were shown on screen only, not spoken.

2025-03-11 · open the original ↗

• Copilot just shortens the search, Stack Overflow, copy-paste loop. Conery, who works on developer outreach for Microsoft's VS Code team, answers the "garbage code" complaint by noting that Stack Overflow answers need the same human vetting; trouble starts when Copilot is asked to do too much. Giard adds his own view that imperfect code getting you 80% of the way is still a big saving.
• Custom instructions pre-prompt every request and shrink your prompts. A markdown file of instructions in a .github folder is prepended to every request in edits, chat and inline chat. Conery keeps it a terse bulleted list of standards (two spaces, brace placement, class naming) and adds example code in fenced blocks, such as a Sequelize user class or nested test blocks, which Copilot follows; output stays non-deterministic. Detail moved into the instructions means a prompt can be as short as "make the data access layer".
• Edits plus strong instructions get about 95% there, in small steps. Edits changes files across a working set and shows per-change diffs to accept or reject. In his demo it turned a five-item spec into a SQL schema, which he then used itself as an instruction to generate a Sequelize data access layer "nearly perfect", missing only a file reference or two. He credits his boss Burke Holland with the rule of very small steps, and admits he could retry his demo until it worked.
• Agent mode runs steps in series and corrects its own errors. Then available only in VS Code Insiders, which updates as often as twice a day, it took his prompt to initialise a Node project, install packages and create a GitHub repo with the GitHub CLI, pausing for "Continue" at each step. He says it installs a missing CLI itself and fixes compiler errors in typed languages such as C# or TypeScript; he had no release timeline.
• Next edit suggestions propagate a change through the rest of the code. Renaming a Point class to Point3D prompts a diff adding a Z coordinate and updating later calculations; Holland's video shows it carrying a string changed to an enum through C# code. Conery was unsure whether it had reached regular VS Code yet.

2025-03-11 · open the original ↗

• Use AI to make yourself more agentic, not to build agents. Godbole, recently retired from Google, argues that once knowledge is commoditized the differentiator is how wise, pro-social and self-authored a person is. He wants AI to support metacognition and a reciprocal loop: acting more, hitting one's limiting beliefs, and growing. He credits Toronto professor John Vervaeke's work on wisdom and relevance as a major influence.
• AI predicted breast cancer from mammograms better than radiologists. Godbole was co-first author of the 2020 Nature paper showing a deep learning system predicted biopsy-confirmed cancers two to three years ahead, depending on the country, on a US and UK dataset of what he recalls as a hundred-and-some thousand patients, and beat every radiologist in a separate reader study. He sees the bigger promise in combining human and AI strengths and in expanding access where patients far outnumber radiologists.
• An LLM must be told which "cinematic universe" it is in. Godbole argues that for an LLM there is no objective fact: pre-training fits the union of all universes in web text, and post-training narrows it to an assistant-and-user universe, so models are statistical models role-playing as human raters role-playing as AI assistants. This first half of his prompting playbook, he says, will outlast its prompt examples.
• Prompt as explicitly as you would brief a stranger. He advises imagining a person off the street with all human knowledge but no knowledge of you, and pasting one's complaints about a failing prompt into the prompt itself. Skills from couples therapy and the book Nonviolent Communication carried over, and he reports an anecdotal inverse correlation between ML expertise and prompting skill.
• Cheaper, better models will unbundle job roles and squeeze undifferentiated products. Extrapolating modestly from current curves, Godbole expects "full-stack employees" such as PMs who arrive with prototypes and user research before engineers are involved. Because copying an undifferentiated product will be cheap, he speculates software could come to resemble the music industry's winner-take-most economics, with discontinuous effects such as small teams not hiring PMs.

2025-03-11 · open the original ↗

• Distilled DeepSeek R1 models run locally on a Copilot+ PC's NPU. Hanselman uses the free AI Toolkit extension for Visual Studio Code on a 12-core Snapdragon Elite laptop with 32 GB of RAM. It offers NPU-optimized distillations of DeepSeek's large mixture-of-experts model at 1.5, 7 and 14 billion parameters, which he calls the "squished", "JPEGy" student versions.
• The 14B model works fully offline, but slowly and inaccurately. He switches the laptop to airplane mode mid-answer to show nothing is going to the web, and reports about 40 tokens a second. The model "yaps" through its thinking and gets details about him wrong, so he warns against using it as a fact machine. The 1.5B model runs much faster and loads the machine far less, at a cost in accuracy and breadth.
• Loading the big model took trial and error. The 14B model first failed to load, which he blamed on having too many models in memory, so he killed the toolkit's inference service agent; he later found the toolkit's own "Unload" option. He notes it also offers a playground, a prompt builder, local fine-tuning and free cloud models through GitHub Models, none of which he tried.

2025-03-10 · open the original ↗

• Record LNG exports are pushing US gas prices up. He cites a report of US natural gas at more than a two-year high on record LNG exports and supply concerns, says February 2025 exports were the largest in history, and notes planned projects are likely to triple them.
• Renewables-heavy Europe is what is driving demand. He argues the world must choose gas or coal because renewables are not cutting it; the top buyers of US LNG, a $25 billion-a-year market, are led by the Netherlands, France and the UK, which he says have the most renewable energy per capita.
• He expects natural gas prices to keep climbing long term. Showing the past year's chart, he says there will be dips and rises, but the long-term trend will be upward as export capacity triples.
• Utilities are using gas prices to inflate delivery charges. He points to $400 to $500 delivery charges on bills, noting $10 gas in 2022 did not bring the bills seen in 2024 and early 2025, and says the fees fund utilities' renewable build-out. He advises planning long term to cut utility costs because it will get much worse.

2025-03-10 · open the original ↗

• This post includes paid content that is not available publicly. See the original source.

2025-03-10 · open the original ↗

• DeepSeek erased the frontier labs' moat, not merely their benchmark lead. Burkov calls it probably the most important event for language models since ChatGPT: a state-of-the-art model trained for what he puts at 5 to 10 million rather than the 20 to 50 million previously assumed, with a published step-by-step recipe that several independent R1-Zero and R1 implementations have already reproduced.
• It is also far cheaper to run. Burkov compares OpenAI's o1 at $60 per million output tokens with DeepSeek's $2, roughly a 30-fold reduction, so that anyone, "your grandmother", can now run state-of-the-art AI.
• Automated verification replaced costly expert-written training data. Burkov argues OpenAI and Anthropic invested heavily, by his estimate billions, in subject-matter experts writing problem-solving dialogues, whereas R1 used binary rewards from checks: does the code compile or produce the right output, does the math answer equal the known value, does the logic puzzle name the right culprit. Hiding a class in a GitHub repo and rewarding a compiling rewrite is one way he says close to a million examples can be generated with no experts.
• Open-weight models are not open source without their data. Burkov's stricter definition, unlike Yann LeCun's, says a model is like a binary: reproducing it requires the training data. The few models released with full datasets are no longer competitive, since today's pre-training uses roughly 15 to 20 trillion tokens versus perhaps four or five trillion in open datasets.

2025-03-10 · open the original ↗

• Colour-code your calendar by how each meeting will feel. Hanselman uses Outlook categories: green for positive personal time such as lunches, red for meetings with bosses or VPs, orange for presentations and podcasts as "approaching stress", pinks and purples for one-on-ones, and blue for general business. He argues a single-colour calendar is like playing Tetris with one colour.
• Grey focus blocks protect time without looking like appointments. He colours focus blocks grey, the same as the calendar background, so the time is held against bookings but doesn't read as a meeting.

2025-03-09 · open the original ↗

• The car-loan interest deduction only helps people who itemize. Answering a pro-Trump clip praising Trump's plan to make interest on American-made car loans deductible as a win for working people, he notes anyone taking the standard deduction gets nothing from it.
• Few working-class taxpayers itemize; top earners dominate. He cites only 1.2% of the bottom 20% of earners itemizing, 2.5% of the 20-40% bracket and 5% of the middle, about 13% overall, with his chart showing 91.5% for the top 1% and 72% for the top 5%.
• Those who itemize largely already deduct vehicle costs. He argues itemizers are mostly self-employed and already write off mileage and maintenance, and many top earners pay cash for cars, so the new deduction adds little and does not reach blue-collar workers.

2025-03-09 · open the original ↗

• To eliminate hallucination, don't let the LLM generate answers. Burkov says RAG cuts hallucination sharply, by his rough figures from perhaps 20–30% of answers to 1–2%, but not to zero. For his company's closed-domain labor-market product, the team instead fills predefined templates ("the average salary for a nurse in San Francisco is…") with numbers pulled straight from internal APIs.
• Showing the system's interpretation makes remaining errors visible. The LLM only parses the user's string into structured fields, which are normalized to internal taxonomies (e.g. "JS" to JavaScript) and displayed as labels before any number appears. Errors can still come from interpretation or imperfect scraped data, but Burkov says the chatbot's answer matches what the conventional UI would show; the LLM is used "to understand but not to inform".
• Bolted-on uncertainty warnings can be overcautious. The host reports that Claude now often says a topic is niche and should be double-checked, including on questions he doesn't consider niche, which he takes as a layer on top of the core model that still needs tuning.

2025-03-09 · open the original ↗

• Knowing a language's design culture helps you trust it. Torgersen, lead designer of C# at Microsoft (visiting in 1999, joining in 2005), says users choosing a language should look at how much it evolves, how open it is and its track record on developers' problems, as with any vendor.
• C# must keep evolving or become legacy, without ever breaking code. He calls the team "on the aggressive side" of language evolution, adding large features when warranted even though nothing can be removed. With billions of lines of C# in use, much of each feature's work is fitting it in without breaking customers or clashing with existing features, while trying not to follow fads.
• Designing in the open is overwhelmingly positive, despite visible heartbreak. He calls the change "98% positive". Ideas arrive mainly as user-reported problems on the language-design GitHub repo and accumulate until the signal justifies investment. Users watch features get prototyped then shelved, but he says openness makes the conversation human-to-human; Giard's version is that it is easy to hate Microsoft but hard to hate Mads.
• Competing languages are healthy competition and a source of ideas. He would only stop evolving C# if it "can't turn this next corner"; so far it has, including trimming ceremony in response to newer languages. Borrowing, such as index and range support after Python's easy indexing, "borders on stealing but we all do it".
• His proudest contribution is async/await. He drove the feature to end error-prone asynchronous programming, and is most proud that many other languages adopted it.

2025-03-09 · open the original ↗

• US oil imports are a business, not a matter of need. He argues companies import crude because they profit from refining and selling the products; no government official is setting imports by comparing production with consumption, so how much the US produces has nothing to do with it.
• All recent US production growth has been exported. He says 100% of the increase in US oil output since 2016 has left the country, and the US already produces more than 10 quadrillion BTUs of energy beyond what it consumes.
• More drilling will not reduce imports by a drop. Even doubling total oil production would leave imports unchanged, he argues; any Trump-era increase, which he does not expect in the first year, would be exported as under Biden and Trump's first term. He calls "energy dominance" and self-sufficiency talk from either party nonsense.

2025-03-08 · open the original ↗

• Hydrocarbon chain length determines what each product is. He explains that shorter carbon chains make lighter compounds: the shortest, around two carbon atoms by his account, are natural gas, then gasoline at around four, then diesel, then heavy oils. Refining is breaking these chains down and putting them back together.
• Crude is first boiled to vapor at a gravity-set temperature. The crude goes into a boiler and is heated until it vaporizes; heavier crude, with a higher boiling point, must be heated more, and lighter crude less.
• The distillation tower sorts the vapors by boiling point. The tower is hottest at the bottom and coolest at the top; rising vapors condense on its plates as they reach their boiling points, yielding heavy oils at the bottom, then lubricating oils, gas oil, kerosene, gasoline, naphtha and gas at the top. This thermal fractional distillation is the first step before further processing.

2025-03-08 · open the original ↗

• Trump's offer of fast-track citizenship reverses his long-held immigration record. He cites over 107 Trump tweets containing "aliens", 10 on birthright citizenship, an executive order trying to end birthright citizenship, and administration plans to deport 10 million people to Mexico, Central and South America, with some sent to Guantanamo Bay.
• He reads race as the only explanation for the reversal. The beneficiaries are white South African refugees, whom he contrasts with Haitians Trump accused of eating pets. They face no $5 million gold card or IQ test. He says he does not oppose admitting genuine refugees, but calls the double standard glaring and sarcastically congratulates the president for "evolving".

2025-03-08 · open the original ↗

• This post includes paid content that is not available publicly. See the original source.

2025-03-08 · open the original ↗

• AGI is possible but as speculative as teleportation. Burkov declines to give a timeline, saying only that since evolution produced human minds, something analogous might be built.
• Infinite-horizon planning is what LLMs, and other animals, lack. He contrasts crows that use cars to crack nuts and apes that keep a stick overnight but discard it after three or four days without bananas, with humans who plan for generations, as with climate change. LLMs are reactive: an "agent's" agency comes from its system prompt, not itself, so it can be violated unpredictably. The host adds his own hypothesis that AGI may need brain-like modules, such as a prefrontal-cortex analogue, rather than a scaled-up single architecture.
• LLMs cannot tell what they don't know. Burkov asked OpenAI's o3-mini whether his new book, The Hundred-Page Language Models Book, released after its training cutoff, is good; its visible chain of thought admitted knowing nothing about it, yet the answer confidently said experts and readers praise it. He says a Qwen model fine-tuned to answer "I don't know" about obscure people denied knowing him, then named him as author of his earlier book, showing such fixes are patches, not self-knowledge.

2025-03-07 · open the original ↗

• Cutting off Ukraine predictably emboldened Russia. He argues that after weeks of Trump vilifying Zelensky and propping up Putin, the White House blow-up and the halt to all US aid, Russia sharply ramped up attacks and struck and sank a Swiss-owned civilian cargo ship; Trump is now surprised and threatening more sanctions.
• Trump lacks the knowledge and humility to out-negotiate Putin. He says a real estate developer who won't learn the history or heed advisers is outmatched by someone who has done this his whole life, and that chest-pounding and flag-waving have never produced a peace deal, whatever Trump says about wanting fewer lives lost.

2025-03-07 · open the original ↗

• The US is almost wholly dependent on Canadian potash. Responding to another creator who raised the issue, he says the US gets something like 95% of its potash, the base of fertilizer, from Canada, whose share of world supply he compares to OPEC's share of oil.
• Ethanol from fertilized corn holds down all gasoline prices. He says ethanol adds about a million barrels a day to fuel supplies, lowering pump prices even for ethanol-free gasoline, though its all-in cost is probably higher than regular gasoline's.
• A potash cutoff would hit fuel before food. Alternatives are Russia, China or Belarus, none of them friendly; if fertilizer ran short, he expects food to take priority and the US to lose that million barrels a day, pushing ethanol and non-ethanol fuel prices up significantly unless supply were found elsewhere.
• Potash may be a stronger trade weapon than Canadian oil. Not being a potash expert, he guesses Canada could sell it to more markets than its oil and so weather a trade war, though he thinks the trade war is all but dead and "just a game."

2025-03-07 · open the original ↗

• dbt's real value is documented, agreed business logic. In a clip from episode 861, a former pro athlete turned data engineer says dbt let her team embed business logic agreed with stakeholders, with field definitions (dbt docs, now auto-generated, plus reusable doc blocks) that force agreement on what terms like "active customer" mean and help new engineers. She describes Sigma, sitting on Snowflake, as easing adoption for Excel-minded PMs.
• LLM prompts need a compiler the way HTML needed React. Gupta, CEO of Boundary (episode 859), says BAML ("basic-ass machine learning") was inspired by React catching an intern's missing closing div that could break a news feed. Because every codebase will hold more prompts each year, he argues developers consuming models need type-safe tooling with instant error feedback, as TypeScript gave JavaScript.
• A tabular foundation model is spreading across scientific applications. Hutter (episode 863) and the host note v1 was already used in a Science paper on chemical proteomics, and an "awesome TabPFN" repo Hutter had just created lists about a dozen applications, including insurance, fault classification, finance, wildfire propagation and algal blooms. Version 2 handles more data types, missing data, outliers and larger datasets, and can be fine-tuned for domains its authors know nothing about.
• A mandatory paid discovery phase made Pandata scalable. Al-Dhubaib (episode 865) says returning clients valued plain speaking and a 30-day to six-week "discovery and design" engagement, so he made it compulsory, priced then at 12,000 and now 50,000. The pipeline briefly dried up, one of three times Pandata's bank account fell below a month's expenses, but he then landed three of his biggest clients, two of whom stayed about six years until the exit; his lesson is the art of saying no.

2025-03-07 · open the original ↗

• A migration is far harder than copying files. Beaudoin-Allaire, a principal software engineer on ShareGate at Workleap, says you don't own the data or know what you'll find: a thousand documents or a million, odd settings, and sources and destinations with different capabilities. Large files must be streamed from source to destination rather than held in memory, and Microsoft 365 now has chunked upload sessions for them. He recalls the file-size limit growing from 2 GB to perhaps 50 GB or more.
• Microsoft Graph simplified things, but legacy APIs are still needed. Graph gives one way to authenticate and consent, documented calls, and a dependable SDK, unlike older SOAP and CSOM calls that were barely documented. It still has blind spots, so ShareGate keeps an ordered fallback list that includes SOAP calls and exposed SharePoint DLLs, while on-premises versions from 2003 onward need their own APIs. Hanselman calls Graph one of the smartest things Microsoft 365 ever did.
• Every extra call is multiplied a millionfold. Adding one call to the ten or so needed per file version adds up across a million files and can trigger throttling, which he likens to a car braking for no reason and slowing all the traffic behind it. Limits differ by service, shared in SharePoint and per mailbox in Exchange, so work is parallelized to match, and the software is built around fault tolerance in the spirit of chaos engineering.
• ShareGate exposes a performance slider rather than adapting automatically. Beaudoin-Allaire says they cannot tell how close a tenant is to its limits, and must handle both graceful 429 "too many requests" responses and 503s, the service saying it is "on fire". The slider predates cloud rate limiting: on-premises servers could crash under load before signalling anything. Most of the time the network, not the customer's machine, is the bottleneck.
• Metadata that doesn't map gets an approximation and a warning. Permission levels differ between Google and Microsoft, so ShareGate maps them to read and write equivalents and lets customers remap them. Where something cannot be supported, it says so in documentation or warns during the copy. Google Drive files appearing in several places through links were copied to one place, with a warning.

2025-03-07 · open the original ↗

• A blog platform must make leaving easy and make content last. Reece, founder and CEO of Micro.blog, says the service archives pages you link to, keeps a paid blog online unless you ask otherwise, can ping the Internet Archive on every post, and mirrors posts and photos to GitHub about once a week. He says the platform would be "hypocrites" to preach the open web while trapping content, which Hanselman ties to the link rot he finds on working actors' expired sites.
• Micro.blog deliberately stays small. Reece argues that very large, ad-based platforms develop bad incentives and moderation problems, while small fediverse communities can manage themselves. He says tens of thousands of users is fine, and that at 100,000 they would probably have to stop taking new users. He took Hanselman's word "boutique" for it.
• Discovery is hand-curated and shows no follower counts. Reece says trending hashtags let communities clash and invite hateful piggybacking, so a human curates the Discover page from recent posts. An early "people to follow" list was dropped because it snowballed popularity, as Hanselman recalls Twitter's suggested-user list doing. Reece concedes the balance may lean too far toward making people hard to find, and is open to linking distributed starter packs.
• POSSE concedes that not everyone reads RSS. Publishing on your own site and syndicating elsewhere lets you keep control while still reaching people on social networks. Reece says he added cross-posting first for Twitter, because existing tools added unwanted links and he wanted short posts to read like tweets. It now supports eight or nine services, each coded to its API, though platforms such as Instagram lock down posting.
• A $1-a-month plan is an experiment made possible by static hosting. Micro.one launched a week before the recording with domain, photo and podcast hosting, alongside the $5 standard plan, which Reece says will always be $5, and a $10 premium plan. A credit card is required, mainly to deter spammers and AI-generated content farms, and unlimited photo storage is his main cost worry. Blogs are published with Hugo as static HTML served separately from the platform, so they stay up during an outage.

2025-03-07 · open the original ↗

• Trump's trillion-dollar Saudi investment figure is not plausible. In the clip Trump says Saudi Arabia will invest at least $600 billion in the US and he will ask the Crown Prince to round it to about $1 trillion. He counters that $1 trillion is three times Saudi Arabia's total annual budget, the equivalent of the US pledging $15 trillion to Saudi Arabia.
• He reads the repeated trillion-dollar deals as a pattern of inflated claims. He cites the trillion-dollar LNG deal with Japan, which he says would take 650 years of Japanese LNG purchases at current prices, and the trillion-dollar Ukraine minerals deal, and faults the public for accepting each without challenge.

2025-03-07 · open the original ↗

• Visual Studio on Windows has a built-in remote Linux file explorer. Hanselman shows that adding the "Remote File Explorer for Linux" component, about 30 MB under the Linux and embedded development workload in the Visual Studio Installer, puts a Remote File Explorer on the View menu. With it he browses his PiDP-11, a PDP-11 replica kit running on a Raspberry Pi, opens its start-up shell script, has Copilot explain it line by line, edits it with Copilot completions, and saves it back to the Linux machine.
• Visual Studio's terminal can SSH straight into the remote machine. The terminal's connection list includes his WSL distributions and the remote connection, so he opens a shell on the PiDP-11 without any setup and boots Unix V7 on it. In the same terminal he also runs YEdit and htop, which he takes as a sign of how much the team cares.

2025-03-06 · open the original ↗

• The commentator's USMCA defense contradicts his own earlier claim. In the clip, a pro-Trump commentator says Trump scrapped NAFTA in 2018 for USMCA so Canada and Mexico could no longer take advantage; he notes the same person said two hours earlier that both countries charge the US exorbitant tariffs, and asks why tariffs are needed if USMCA fixed it.
• Trump did attack Canada's sovereignty. Against the commentator's claim that no one questioned other nations' sovereignty, he says one of Trump's first moves was telling "Governor Trudeau" Canada would become the 51st state, which means ending it as a sovereign nation.
• Trump has disowned the trade deal he signed. He points to Trump calling it the most horrible trade deal he had ever seen, despite having signed USMCA and taken credit for it.

2025-03-06 · open the original ↗

• The Mexico tariff delay effectively exempts nearly everything. After US automakers won an exemption the day before, Trump exempted all goods under USMCA with Mexico for another month, as he predicted; he notes USMCA covers almost everything, and reads it as Trump realizing tariffs hurt the economy, suggesting he stop listening to Peter Navarro.
• The tariffs' purpose is unclear, and their defenders are exposed. He speculates about a negotiating tactic, a news-cycle distraction, a show of dominance, or a ploy to sink the stock market for a quick profit, and asks Republicans why, if tariffs are so good, they keep getting delayed as the economy weakens.
• First-quarter GDP outlook remains deeply negative after revision. He reports the Fed's forecast improved from a 2.8% contraction to 2.4%, still really bad; two quarters at that pace would mean a deep recession. He also cites a Gallup poll putting trust in media at an all-time low of 31%.

2025-03-06 · open the original ↗

• Nationalization would require crashing prices, and only OPEC would cooperate. Responding to a question about using low OPEC prices and "drill baby drill" to break the private sector, he agrees that would be the only route: drive prices low enough to bankrupt companies, then buy everything. US producers would never join in driving the price down.
• The US cannot afford to buy its oil and gas industry. Privately held acreage alone is worth roughly $2 trillion, paid up front, plus an agency of at least 50,000 people to manage it, some of which could fold into the BLM. Owning the upstream, midstream, marketing and retail chain adds about another $3 trillion; selling minority shares, as with Saudi Aramco, would cover a tiny fraction.
• The investment would never pay back if the goal is cheaper energy. He estimates any return would take well over 50 years, by which point a national industry would no longer be needed. A cheaper route, letting federal leases expire and running alongside private firms, he suspects would not work well.
• Nationalization only works early in a country's industrialization. Doing it after becoming the world's largest energy producer is, in his analogy, like buying Tesla stock at its peak and expecting to get rich: you lose money.

2025-03-06 · open the original ↗

• The Canadian offer gave the president everything rather than defying him. In the clip Danielle Smith proposes doubling Canadian crude moving south, promising jobs and lower fuel prices. He says Canadians praising her for telling Trump off have it wrong, the "we'll sell it elsewhere" threat is empty with no route to other markets, and he suspects she wants Trump to take Keystone XL.
• Doubling Canadian imports would make Canada the sole US supplier. The US imports oil from about 46 countries, and he argues it lacks the capacity and storage to double Canadian volumes and keep the rest. Relying on one trading partner for such an important resource is, in his view, unwise even when it is Canada.
• The jobs and price benefits are overstated. The infrastructure would create construction jobs, but pipelines employ few people once open, and he dismisses the promise of lower prices for Americans as "not a thing."

2025-03-06 · open the original ↗

• The Looneys came to games from NASA engineering. They met at NASA in 1986: Kristin designed VLSI chips starting as a co-op student in 1985, and Andy wrote telemetry software, a file-transfer tool for a real-time OS, and later code for a Hubble coprocessor. Andy, whose father helped found Goddard in 1959, stayed eight years; Kristin, a little longer.
• Fluxx exists because they could not afford pyramid molds. Custom injection molds would have cost $10,000 or more, so Kristin asked for a card game; Andy wrote the memo describing it on 24 July 1996. They say it has run 30 years, sold four and a half million copies, spans about 45–46 versions, and is what "makes us all of our money".
• Andy treats game rules as software run on humans. Each card is like code that supersedes others, play starts with "draw one, play one", and themed versions forced new card types: creepers for Zombie Fluxx, surprises with in-turn and out-of-turn clauses so every card stays playable, the secret in the upcoming Mystery Fluxx. Word Fluxx has 51 keepers against the usual 22, and Solo Fluxx, cracked after 20 years, meant retrofitting every edition to one to six players.
• Luck is deliberate, so anyone can win until the end. They say the balance lets geeks and their non-gamer families play together, at the cost of hardcore gamers' dislike. They also tell fans loudly when only packaging changes, as with Monty Python Fluxx's licensor-required box, rather than manufacture collectible versions.
• The pyramids are a game system rather than one game. Pyramid Arcade bundles 22 games: 20 by Andy, Kristin's Volcano, and Homeworlds by Andy's childhood friend John Cooper, a no-luck strategy game Andy says rivals chess, with a medal for beating him awarded 21 times. Fans have made hundreds of pyramid games, foreign publishers license translations (Fluxx is in at least a dozen languages), and several titles play online at Board Game Arena.

2025-03-06 · open the original ↗

• Executives' agent hype reflects their shareholders' interests. Against the host's list of boosters, Burkov says Nvidia's CEO talks up agents because it implies buying more GPUs, and that Salesforce once declared traditional software dead in favour of SaaS, which did not happen, and now says the same of SaaS. He distinguishes Andrej Karpathy as careful to describe possibilities rather than predict them.
• A single agent is a rebranded LLM, useful only in-distribution. Burkov's best case for agents is gathering and summarizing web information, because web data is closest to what LLMs were trained on; he questions how big that market is, likening some of it to a cron job running Google searches. Dropped into a company's intranet of legacy systems they have never seen, he expects agents to "break their teeth" quickly.
• LLM-based multi-agent systems are effectively undebuggable. Burkov, whose PhD was in multi-agent systems, explains that distributed systems are already hard to debug because you cannot pause all asynchronous actors at once, and LLM agents add black-box internals you cannot step through. Reliable multi-agent systems need full control of each actor and inspectable communication protocols.
• Agents need human-level trustworthiness before being combined. Burkov argues that even human systems guard against unreliability with redundancy, such as two keys turned far apart, and that today's agents are nowhere near as reliable as a person, so multi-agent systems built on them are "a recipe for disaster". He says no one has a clear idea how to reach a trustworthy AGI.

2025-03-05 · open the original ↗

• Saying America needs no Canadian oil undercuts years of pipeline complaints. Addressing Trump sarcastically on his claim that the US does not need Canadian oil, he asks why Trump, Ted Cruz, Jim Jordan and Lindsey Graham spent four years saying the Keystone XL cancellation drove gas to $5 a gallon. He welcomes Trump to his side and asks for it in writing on White House letterhead.
• Dropping Canadian oil means buying from hostile producers instead. The replacement would be OPEC, and he mocks the idea of replacing the closest US ally with nations he describes as funding terrorism, and suggests Republican voters would pay more for gas if told it meant owning Canada.
• Earlier Russian oil purchases helped fund the invasion of Ukraine. He says the US bought a lot of Russian oil while Trump was president, and tens of billions of dollars of it in the couple of years before the invasion, then cut it off without harm, proving it was never needed. His sarcastic reading is that the US financed Russia's invasion and then Ukraine's defense.

2025-03-05 · open the original ↗

• Canada's dependence on US trade mirrors America's on China. Answering a video arguing Canada needs the US, he grants that 75% of Canada's exports go to the US but says the US's own biggest trading partner is China, which it calls its biggest enemy.
• The video's energy and defense complaints don't hold up. He mocks the claim that the US "allows" Canada to sell it discounted oil, notes the video blames Trudeau for shutting down Keystone XL when Republicans usually blame Biden, and answers its point that Canada spends 1.38% of GDP on defense against 3.8% for the US by saying America chooses to spend nearly $1 trillion a year.
• Anti-Canada sentiment was manufactured by Trump. He argues followers have adopted Trump's hostility wholesale, when Trump's insults, such as calling Trudeau a governor, are a negotiating tactic he uses because his base enjoys it rather than something he means.

2025-03-05 · open the original ↗

• Systems thinking matters more as AI does the coding. Hanselman relays a systems architect's argument, from what he calls his most popular Hanselminutes episode, that the learn-to-code movement taught for-loops but not the systems code lives in. Her example is a cheap Amazon package that arrives next day with nobody thinking about the oil, freight, labour and macroeconomics behind it. Mark agrees and equates systems thinking with architectural thinking, saying that with AI writing so much code, the value lies in the thinking on top of it.
• Sketch the rough shape, then sculpt iteratively. Mark says he starts a Sysinternals feature from a broad structure and reworks it as he goes, likening coding to sculpture where the chisel strikes can't be predicted. Hanselman says his own systems thinking leaves him stuck among nine ways to do something. In brownfield code, both say you should sculpt in the previous sculptor's style rather than bolt things on.
• Nobody holds today's systems whole, so learn just enough to decide. Mark relays Scott Guthrie being asked who still has Azure in their head; the answer was nobody, and Mark says he knew a far larger share of Azure 10 to 15 years ago. As a senior executive reviewing experts' work, he asks how much he needs to know to make a call, looks for risk areas from experience, and asks questions where there are gaps.
• Mark sees "no line of sight" to AI replacing programmers. His first reason is specification: you never fully specify what you want, the AI's fixes break other things, and you end up owning code you didn't write, which leaves a non-coder "dead in the water". Hanselman's example is a site built with his 17-year-old, where the AI made red ones-and-zeros "Matrix rain" from 256 hard-coded spans that overheated his laptop. Mark says over 95% of his own research code is AI-generated, which he can manage only because he understands it.
• AI is stuck on old APIs and thin documentation. Mark says a model generated code against an older OpenAI SDK because it had seen more examples of it, and that generated reference docs lack the context a human or an AI would need to use a new API. His conclusion is that AI accelerates experts but limits beginners to simple projects, so young people still have to learn systems and how to program.

2025-03-05 · open the original ↗

• This post includes paid content that is not available publicly. See the original source.

2025-03-05 · open the original ↗

• TalentNeuron turns scraped job postings into workforce insights with NLP. The host cites Burkov as having previously said the company makes 95 million daily predictions from 700 million job-posting data points. Burkov describes some 35,000–40,000 rule-based robots downloading postings, then models that extract and normalize skills, telling "excel in Word" from the skill Excel, and predicting industries with BERT and RoBERTa since around 2018–2019.
• Multilingual models removed the need to label data per language. Where the team once built separate salary extractors for about 25 languages, Burkov says they now reuse labeled data from other languages to cover new countries, a capability he says he would have called impossible ten years ago.
• LLMs now build about half of the company's scrapers automatically. Given a corporate home URL, the system finds the careers section and writes extraction rules for titles, locations, dates and descriptions; the remaining harder half is still done by engineers.
• Think hard before putting an LLM in front of users. Burkov tells decision-makers to think "30 times", warning that LLMs work in-distribution but businesses cannot know what is in-distribution because training data is hidden, even for open-weight models. Systems that pass developers' tests can fail arbitrarily on real production data, with no way to detect the failure or know whether more fine-tuning examples will fix it, costing reputation and angry customers.

2025-03-05 · open the original ↗

• Go slow to go fast: treat data deliverables as products. Justin Chen, a former head of data science at Coinbase who now leads growth data science at Mercury Insurance, says quick ad hoc SQL creates rework and siloed knowledge once a company scales. Coinbase worked that way at about 800 people, but as it grew to 1,000–2,000, one person's quick query in Airflow became the core table for downstream reporting and failed daily. His fix was a tech design doc for any work beyond six hours, taking two to three days, plus mini-milestones and small wins instead of six-month projects.
• The right rigor depends on the product's phase and domain. Before Series A, product intuition matters more than data. At product-market fit, early signals matter, and scaling (say, 100 to 10,000 customers) is when to build reusable solutions. The same logic applies to products inside a large company. Growth teams should launch fast and fail fast, platform teams must weigh compliance risk, and B2B teams release slowly, as LinkedIn Talent Solutions did with batched releases.
• Unicorn data scientists become bottlenecks; build systems instead. Very capable generalists get every hard problem, document little, leave gaps when they go, and can resist change, he argues. Leaders should instead provide onboarding, tooling and training so anyone who clears the bar succeeds. Coinbase's process had a rigorously graded presentation and two technical rounds, but he treats evaluation as continuous after hiring. He stresses technical chops over an early switch to management, and learning over pay early in a career. A host's view, which Justin endorses, is that 7/10 across skills plus a 9/10 "superpower" that fits the role beats 8/10 everywhere.
• AI will replace ad hoc, reactive analytics work. In 2023, with no headcount, junior data engineers at Coinbase built a GPT-3.5 Slack bot. It answered metric questions from the semantic layer, made charts, returned metric definitions, and could kick off Airflow backfills. He argues that necessity drives this, since stakeholders with unlimited analysts would rather ask a person and share the blame. A host calls it replacing a data scientist. Justin calls it "the trend itself" and says leaders must also equip employees for it. He expects semantic layers and clean metadata to matter more, and stakeholder trust built through Slack requests to need rebuilding.
• Data teams should act as partners, not support. Rather than handing over a metric or a large spreadsheet, deliver an interpretation tied to cost, revenue or user experience. Treat communication as more than half the work, and build solutions teammates can reuse, such as an experimentation platform instead of a manual process.

2025-03-04 · open the original ↗

• This post includes paid content that is not available publicly. See the original source.

2025-03-04 · open the original ↗

• LLMs are for prototyping; production still needs trained, controlled models. Burkov, ML lead at TalentNeuron, warns that businesses cannot know whether their problem is in-distribution because training data is hidden, so systems that pass tests fail unpredictably on real users. His four-person team uses instructed LLMs as placeholders at the many decision points in a pipeline, then replaces the weakest links with fine-tuned or purpose-built classifiers, because errors such as 15% at several steps compound toward zero end-to-end success.
• Agents won't work in real life, and multi-agent systems can't be debugged. Burkov, whose PhD was in multi-agent systems, calls a single agent a rebranded LLM with the same out-of-distribution problems, useful mainly for web information gathering. Asynchronous black-box agents cannot be paused and stepped through, so he calls building multi-agent systems on them "a recipe for disaster" until agents are as trustworthy as humans. He attributes CEO hype from Nvidia and Salesforce to shareholder interest.
• LLMs lack infinite-horizon planning and a sense of what they don't know. Burkov likens AGI's prospects to teleportation and declines a timeline. Humans plan for generations where apes discard a tool after three or four days, while LLM agency comes only from the system prompt. His test of OpenAI's o3-mini on his new book, released after its cutoff, produced a reasoning trace admitting ignorance but a confident answer saying experts praise it.
• Closed-domain chatbots avoid hallucination by keeping LLMs from writing answers. At TalentNeuron the LLM only parses queries into structured fields, which are normalized and shown to the user, while numbers come from internal APIs into fixed templates. Burkov contrasts this with RAG, which by his rough figures cuts hallucination from perhaps 20–30% to 1–2% but not to zero.
• DeepSeek removed the frontier labs' moat. Burkov says it trained a state-of-the-art model for 5 to 10 million instead of 20 to 50 million, published a recipe others have reproduced, and charges $2 per million output tokens against o1's $60. Its R1 training used automatically verified rewards (code compiles, math matches) rather than expensive expert-written data. He adds that open-weight models are not open source without their training data.

2025-03-04 · open the original ↗

• This post includes paid content that is not available publicly. See the original source.

2025-03-03 · open the original ↗

• This post includes paid content that is not available publicly. See the original source.

2025-03-03 · open the original ↗

• This post includes paid content that is not available publicly. See the original source.

2025-03-03 · open the original ↗

• Sell to the buyer where your business creates the most value. Al-Dhubaib says firms had long approached Pandata because buying a data science team is easier than building one. When the head of Further (then Search Discovery), a fellow Case Western alumnus seeking to move into cloud, AI and advanced ML, offered to buy within five minutes of meeting, he first asked about core values to judge whether his clients and team would thrive there.
• Upfront compatibility enabled an unusually fast deal. Al-Dhubaib found the two firms shared problems, attitudes and pace despite differently named values; diligence and paperwork ran from January to March of last year, about three months. He attributes the horror stories he hears about other exits to skipping that matchmaking, partly because it is hard to say no to good-sounding offers.

2025-03-02 · open the original ↗

• This post includes paid content that is not available publicly. See the original source.

2025-03-02 · open the original ↗

• Evaluation design and LLM operations are today's scarce data science skills. Al-Dhubaib says what "data scientist" means has changed drastically over ten years; he now seeks people who can design statistically sound evaluation and quality control for unstructured, multimodal data without leaking what they predict into the data, and ML engineers experienced in deploying language models and diagnosing drift or odd behaviour, which he calls AI Ops.
• The evergreen traits are soft skills, not technical ones. These are explaining plainly, a drive to learn, and comfort with ill-formed problems. Whatever the technical skills of the moment, Al-Dhubaib says these core values from Pandata are what he consistently looks for and recommends data scientists cultivate to stand out.

2025-03-01 · open the original ↗

• Democrats keep failing to back the women who warned them about Trump. After a clip of Kamala Harris saying that with Trump as president Putin would be sitting in Kyiv and that dictators want him back because they can manipulate him with flattery, he says Hillary Clinton also predicted what Trump would do and Democrats didn't turn out for either.
• Far-left leadership and identity politics alienate a more centrist base. He says apart from Biden, whom he calls somewhat centrist, party leaders sit well left of a membership that includes many centrists and some center-right voters, and that catering to groups rather than broadly shared issues is why he left the party to become an independent.

2025-03-01 · open the original ↗

• The Oval Office clash with Zelensky was a betrayal echoing Kremlin lines. Recorded the day of the meeting, the guest, Missouri farmer and political consultant Will Westmoreland, urges comparing Trump's and Vance's points with the Russian UN ambassador's last speech. He calls the claims easily provable lies: US aid was $173 billion, not $300 billion, with perhaps $30–45 billion in cash, and European aid was not all loans.
• Frozen USDA reimbursements leave farmers holding the bill. Farmers fronted hundreds of thousands of dollars for conservation work, such as warm-season grasses as Missouri's grow zones move north, then payments stopped after the inauguration. Westmoreland says only about 1%, roughly $20 million of some $2 billion, has been released, first to the loudest farmers on social media.
• Medicaid cuts would gut rural hospitals and ambulance service. Westmoreland argues the targeted savings cannot be found without Medicaid, which insures 72 million Americans and keeps rural clinics, hospitals and special-needs school programs running. Injured by a tractor power take-off at 12, he lay for 90 minutes awaiting help; today an ambulance arrives in seven.
• Tariffs permanently handed the Chinese farm market to Brazil. Brazil's share of China's row-crop purchases rose from 13% to 43%, and with record pork herds and reserves China is not hurting this time. He says net farm income doubled under Biden, whose agriculture secretary secured record 2024 soybean sales, and that broken H-2A visas leave legal workers arriving 22 days late on average.
• Winning back voters means matching their anger, the guest argues. He puts the "anger index" near 80% and cites a poll in which 52% of Democrats thought Trump better represented working Americans; his new advocacy group and PAC target six close House races for 2026. Kari Odermann adds that correcting a belief takes five to seven repetitions, so audiences must be met where they are.