Jack

I’m a small business owner, also board director, consultant, investor, and father of two. Thanks to AI, I can now build my own productivity tools—helping me drive multiple projects forward without losing my sanity. Runlo - AI Chief of Staff, my main gateway to work and life. getrunlo.com CruxCards - capture structured insights while browsing web, review on mobile while on the move cruxcards.com TrioChart - I am not an active trader but I like to keep an eye on market actions across timeframes triochart.com

personal AI agent is eating apps - first victim might be my own cruxcards lol i am replacing it with hermes with karpathy's llm wiki approach but boy it is expensive - injesting a book costs about $5-8 (opus)!

claude becomes annoying - it is judging, lecturing and preaching. it is no longer like an assistant.

Filter out the AI madness Hot yoga is a perfect treat to your body and mind

have been trying to move 300G from onedrive to google drive claude helped me to use rclone, and suggested to open digitalocean vps for faster transfer fixed quite a few configuration errors along the way, but i do have to manually intervene, such as resizing vps if DO had a cli - claude can complete the whole task by itself

@gnoble79 justified?

Wall Street is WRONG about Oracle. $ORCL is being pitched as the "fourth hyperscaler." The AI infrastructure play of a lifetime. 35 out of 46 analysts have a buy rating. Consensus price target is $246. The stock is at $172. Down 47% from its September high. Now let me explain what the bulls aren't telling you and why this will end HORRIBLY: Oracle's non-current debt has ballooned to $124.7 billion. Up from $85.3 billion a year ago. A 46% increase in 12 months. Total liabilities sit at $206 billion against shareholders' equity of $39 billion. That's a 5-to-1 leverage ratio on a company being pitched as a "safe" infrastructure play. But that $124.7 billion isn't even the full picture... Oracle has been using project financing structures (loans repaid from projected future cashflow) to keep tens of billions more in borrowing off its balance sheet entirely. So when analysts quote Oracle's debt load, they're UNDERSTATING the actual exposure by a meaningful margin. Interest expense jumped 32% YOY. Free cash flow is negative $24.7 billion on a trailing basis. The company is spending $48 billion a year in capex while generating roughly $17 billion in operating cash flow. They issued $43 billion in senior notes in 9 months. They are borrowing at a pace that would make a leveraged buyout firm nervous. And what did they get for all that spending? They fired 30,000 people. On March 31st, Oracle sent an email at 6 AM to tens of thousands of employees telling them their roles were eliminated. 18% of the global workforce gone in a single morning. TD Cowen estimates the layoffs save $8 to $10 billion in annual cash flow. Which tells you everything about the math: Oracle can't fund $50 billion in AI capex AND keep 162,000 people on payroll. So the people went. Net income was up 95% last quarter. The stock is still down 47% from its high. Mr. Market is telling you something. The earnings look great on paper partly because Oracle extended the useful life of its servers to 6 years, reducing depreciation expense by billions. I've been flagging this accounting game across the hyperscalers for months. It flatters the income statement while the balance sheet quietly deteriorates. Now let's talk about the $553 billion in Remaining Performance Obligations that every bull cites as the "reason" to own this stock: Roughly $300 billion of that is a SINGLE contract with OpenAI through the Stargate project. Revenue doesn't start flowing until 2027. And OpenAI itself expects to lose over $167 billion through 2028 even if it hits $100 billion in annual revenue. So Oracle is borrowing $125+ billion to build data centers for a customer that cannot even fund its own operations. And the data centers themselves are significantly behind schedule: The flagship Stargate campus in Abilene has been under construction since mid-2024. 2 years later, only 2 of 8 planned buildings are operational, covering about 200 megawatts of the planned 1.2 gigawatts. The remaining Stargate sites across Wisconsin, New Mexico, Michigan, and other locations are in the earliest stages of development. The total estimated cost to build out Oracle's 7 gigawatts of planned Stargate capacity runs around $340 billion. And lenders are already getting nervous. The Wall Street Journal reported that additional capacity at Abilene originally earmarked for OpenAI ended up going to Microsoft instead - because the banks financing the build were uncomfortable with their credit exposure to OpenAI as the ultimate customer. When your LENDERS don't trust your tenant's ability to pay, then there's SERIOUS issue. And by the time those data centers are fully built, the GPUs inside them will already be approaching obsolescence anyway. Nvidia releases new architectures annually. Each generation delivers dramatically more compute per watt. The hardware goes obsolete in 3 years but the debt used to buy it gets repaid over a much longer horizon. The AI infrastructure buildout is a treadmill, not a revolution. Oracle is the purest expression of that thesis. - $206 billion in reported liabilities. - Billions more hidden off-balance-sheet. - Negative $25 billion in free cash flow. - 30,000 people fired to fund the capex. - A single unprofitable customer behind over half the backlog. - Data centers years behind schedule. And 35 analysts saying buy. This doesn't sound right, does it?

@_The_Prophet__ interesting

⚡️This is the AI substitution showing up in the data for the first time at index level. The 400,000 decline in S&P 500 employment is the inflection point that everyone knew was coming and that almost nobody is ready for. The framing is going to dominate macro debate for the next decade. Most analysts will get it wrong because they will treat it as a cyclical labor cycle when it is actually a structural reorganization of how American capitalism produces value. What the data actually shows. S&P 500 companies cut about 400,000 jobs in 2025, ending eight years of consecutive growth. The companies driving the decline are exactly the ones most exposed to AI substitution and most invested in AI capex. UPS, Oracle, Amazon, Meta, Intel, Microsoft. These are not random cuts. UPS is automating logistics. Oracle is consolidating around AI cloud. Amazon is replacing white-collar middle management with AI tooling. Meta is replacing engineers and product roles where AI now does the work. Intel is restructuring as it loses the chip wars. Microsoft is reorganizing around AI-native operations. The cuts are coming from the same companies that are spending hundreds of billions on AI infrastructure. The capital is shifting from labor to compute. The structural meaning is that the AI capex boom is being funded in significant part by labor cost reduction at the same companies. The companies are not just spending more on AI. They are spending more on AI by spending less on people. This is the substitution happening in real time. The capex shifts to compute. The opex shifts to less labor. The combined math produces higher margins on the same revenue, which is why the AI buildout is being financed without commensurate revenue growth at the AI customers. What this means for the macro picture. Most macro analysts are still operating on the assumption that the labor market and the consumer drive the economy, with corporate profits as the residual. The relationship is inverting. Corporate profits are now driving the economy through capital expenditure on AI, with labor and consumer as residuals that are being squeezed. The squeeze on white-collar labor is just beginning. The 400,000 number for 2025 is the leading edge. The 2026 numbers are going to be larger. Amazon 16,000, Meta 8,000, Microsoft 8,750 in voluntary buyouts that will become layoffs if the buyouts do not hit targets, plus everything that has not been announced yet across the rest of the index. The trajectory is steeper than 2025. The Fed has not understood what is happening. Powell and the FOMC are operating on the standard model where unemployment going up means demand is weakening and the Fed should cut. The current dynamic is that unemployment in white-collar sectors is going up because companies are substituting AI for labor while their revenue and profits keep growing. This is structural displacement, not cyclical demand weakness. Cutting rates does not bring those jobs back. Cutting rates accelerates the AI buildout, which accelerates the displacement. The Fed’s tools are mismatched to the actual phenomenon. They will probably cut anyway in early 2026, partly because Warsh will make the Fed more aggressive on stimulating growth. The cuts will not solve the labor displacement problem. They will pump asset prices and accelerate the AI buildout, which intensifies the structural dynamic.

@petergyang that is why i built my own getrunlo.com

A great personal agent should: 1. Get work done across email, calendar, Google Workspace, or any API/MCP it's hooked up to 2. Act proactively and reliably (e.g., cron jobs, triggers, follow-ups) 3. Have excellent memory that helps it "just get you" over time 4. Work across web and mobile without slash commands or manual setup 5. Let you switch between text, voice, video, and live calling mid-conversation 6. Be reachable from any 3rd party messaging app, just like a real person 7. Have a personality that makes it fun to talk to OpenClaw, Claude Code, Codex - the truth is that none of them check all these boxes yet.

@JayaGup10 accountability is always an issue in corp world

@EHGInvests @jvisserlabs true, hard to chase now - i am looking at $CEG $VST

@sayhellojack @jvisserlabs Intel too!

Stocks completed a record setting rally back to all-time highs. It was led again by semis and the AI capex names. AI remains the driver of the economy and market but during the last month the focus has been on the rapid ascent of Anthropic and their parabolic revenue. In this week's video I go through how their success is driving adoption but is also driving AI to the physical constraints of compute and power. Watch here: youtu.be/sZU5Y0n3YwY

@EHGInvests @jvisserlabs around 27:00 in the video - cards and charts were created by other tools

@sayhellojack @jvisserlabs I didn’t see it :/

@EHGInvests @jvisserlabs in his youtube video

@sayhellojack @jvisserlabs Is this from his paywall?

We are always in a loop of perception, judgment, action, and reflection. The quality of our life depends on how consciously we run that loop.

@pmarca agents become only usable in Q1 26. we will see.

"Surprisingly, occupations with higher exposure to AI have grown faster than least-exposed ones, not slower." Not surprising! Productivity growth -> economic growth -> job growth.

For a dip to look like a true monthly mean-reversion setup, you’d usually want evidence of stabilization first: holding a base for more than a few sessions/weeks, downside volume cooling, and eventually a reclaim of the monthly 5MA area. Right now, the earnings gap argues the opposite: distance from the mean is expanding, not contracting. triochart.com/chat/DlCD6leeXN

@amitisinvesting is after-hour dip a buy opportunity?

service now is more like service down in line with EPS, slight beat on revenues street’s not impressed by it taking all of software down after hours $NOW $IGV

semi is going parabolic

the tool i am building next is training myself on speaking under pressure - high stake conversation if you will it is a skill issue - intense practice is necessary just like any other skill acquisition

@gnoble79 how does consumer sentiment correlate with XLP, which just keeps going up over decades (see monthly)

The S&P 500 just hit an all-time high. But consumer sentiment just hit an all-time low. Both happened in the same week. Think about this. On April 15, the S&P 500 closed above 7,000 for the first time in history. Traders were celebrating on the floor of the New York Stock Exchange. The Nasdaq posted its longest winning streak since 2009 - 12 consecutive days of gains. And 5 days earlier, the University of Michigan Consumer Sentiment Index dropped to 47.6. The LOWEST reading in the survey's 74 year HISTORY. Lower than the 2008 financial crisis. Lower than the COVID crash. Lower than the worst of the post-pandemic inflation surge. Lower than anything recorded since Harry Truman was president. Every single demographic group declined. Every age bracket. Every income level. Every political affiliation. Every component of the index (current conditions, future expectations, personal finances, buying conditions) ALL fell. 1 year inflation expectations spiked to 4.8%, up a full percentage point from March. That's the steepest single-month jump since April 2025. Business expectations crashed 20% in a single month. And 98% of the survey was completed BEFORE the ceasefire was announced. These people aren't pessimistic because of headlines. They're pessimistic because of their grocery bills, their gas prices, and their shrinking purchasing power. Meanwhile, Wall Street is throwing a party. Morgan Stanley's Mike Wilson went on CNBC and declared "the lows are in." He said the market is rotating back into pro-cyclical names and that we should expect things to "resolve constructively." Resolve constructively? The ceasefire expires Wednesday. Iran just re-closed the Strait of Hormuz AGAIN. Brent crude is back near $98. The US Navy is still blockading Iranian ports. Peace talks in Pakistan collapsed over the weekend after Iran refused to abandon its nuclear program. And the stock market is priced for a happy ending. Here's what 45 years on Wall Street have taught me: When markets and consumers disagree this violently, one of them is wrong. The S&P 500 is trading at all-time highs because 7 mega cap tech stocks (powered by AI narratives) account for nearly half the index's market cap. Those stocks run on their own dynamic independent of anything, including the war in Iran. That's a market being dragged to record highs by a handful of names while the real economy deteriorates underneath. Can you call this a healthy market? Software stocks are down 28% this year. The broader SaaS index is down nearly 40%. Private credit is imploding. Gas prices are surging. Consumer buying conditions for big-ticket items are collapsing. But the S&P hit 7,000 so everything must be fine. I remember in 2000, consumer sentiment peaked in January. Two months later the Nasdaq topped. Within a year it had lost 39%. In 2007, the S&P 500 hit its record in October. Consumer sentiment had been falling since July 2007. 3 months later we were in recession. The pattern is always the same: Consumers feel the pain before it shows up in the indices. The market ignores them until it can't.

@amandaorson worth read it again

@mattturck look at the finance, numbers do not lie but if they are pitching as software co, pass by default