- Otherneutrallong-termThe host argues Anduril's laptop is not a PC dabble but a supply-chain-as-arsenal strategy: if you can't build and audit every component, you don't control it.Source
Couch Investor
4 videos
- PLTRneutralshort-termPalantir also reports Monday; the host expects continued strong growth but offers no directional call.Source
- AAPLneutralshort-termApple reports Monday after the close; the host expects broad growth across reporters but gives no specific Apple call.Source
- AI Capexneutralmid-termHe is watching what Shopify says about AI agent shopping, noting companies so far say the impact is still very small.Source
- AMDbullishshort-termHe expects a far more positive AMD report than Intel's. Price at $504, resistance at 526 and 542, daily-200 support at 402.Source
- Consumerpositive for stocksshort-termHe expects a very good Uber report, calling it one of the cheap names with strong free cash flow; autonomous vehicles are not impacting the business yet.Source
- Otherneutralshort-termAppLovin is under pressure; earnings will either prove the market wrong or right — no direction given.Source
- Market Breadthnegative for stocksshort-termSentiment around neocloud names is very bad and good reports are not being rewarded — a cautionary near-term read on IREN and peers.Source
- GOOGLbullishlong-termHe loaded up on Google last year when everyone hated it: Search falling below 50% of revenue reflects Cloud growing 103%, possibly matching Search's size by end of 2027. He would likely buy more if it stays at $280.Source
- GOOGLbullishmid-termAt a 20.2x forward PE for one of the best businesses on the planet, he sees the market's punishment over modest Search weakness as unreasonable.Source
- Otherneutralshort-termRocket Lab barely moved and remains in its main buy zone / accumulation area — a brief restatement of prior positioning.Source
- Otherneutralshort-termSoFi rose but remains in a buying zone at support, with a ceiling just above $17; he flags it as extremely macro-driven.Source
- AMZNneutralshort-termAmazon keeps oscillating above and below $250; nothing has changed.Source
- METAbullishlong-termMeta sits below its 50-weekly ceiling of $630 with a 600–625 buy zone; he calls it extremely undervalued and says every buy debate ends with 'maybe I should just own more Meta' after its ad business crushed the industry.Source
- Otherpositive for stocksmid-termNu's selloff on the Brazilian election is an overreaction in his view; past turbulence never stopped the business, though waiting for earnings is an option.Source
- Otherpositive for stocksshort-termNebius, his largest position, sits in the 148–158 buy zone with a 160 ceiling; he expects a very big move on the November 12 report.Source
- OtherneutralAxon barely moved; not cheap, just less expensive than before.Source
- AVGObullishlong-termBroadcom is still extremely undervalued in his opinion, currently in the buying zone with support near $309, and explicitly on his buy list.Source
- Otherpositive for stocksmid-termCoreWeave looks quite undervalued on paper based on projections, but execution is key; earnings in about two weeks, resistance above $80, support at $50.Source
- Otherpositive for stockslong-termHis PayPal bull spread is well in the green and he still sees money to be made over two years, though he will likely close it to redeploy into beaten-down holdings.Source
- Valuationnegative for stocksmid-termHe fully exited Rubrik after a ~90% gain: a good company but stretched at a 12.1 PEG and ~10x forward sales; even a good report may not be good enough, as Fortinet's 25% crash showed.Source
- Otherneutralmid-termReddit is down 44% in his portfolio; cheap on paper, but he won't add without clarity on a new AI data-licensing deal or user-base stabilization — he'd rather wait for a catalyst.Source
- NFLXneutralmid-termDown 27.5% on a small Netflix position, he is weighing reallocating the $5,400 into something that would probably rise faster as part of year-end consolidation.Source
- MUneutralmid-termMicron at 10x earnings looks cheap, but he worries it could stay cheap forever and become dead money; since it's house money, he's in no rush but keeps debating a swap.Source
- Otherneutralmid-termHe thinks midterms get too much attention — good companies perform regardless, and markets historically rise after midterms, though Iran, inflation and rates make this cycle different.Source
- Fed Policyneutralshort-termPrediction markets imply about a 70% chance of one 25-basis-point rate hike at the December Fed meeting, little changed over the past month.Source
- Otherneutrallong-termThe host argues Anthropic's core issue is not whether partners help it scale, but whether a public company can stay dependent on suppliers who also sit on the other side of the deal.Source
- AI Capexpositive for stockslong-termUsing Amazon as the example, the host argues hyperscaler capex has a visible cash-recovery cycle: 30-year data center assets, servers breaking even in about 3 years, so the spending is not reckless.Source
- AMZNbullishlong-termThe host views Amazon's early free cash flow headwinds as the price of its growth capex cycle, not proof the spending is wrong, implying the cash comes back over time.Source
3 videos
- AI Capexneutralmid-termThe host notes both AI and core businesses are growing rapidly while memory is scarce, forcing real allocation trade-offs between HBM and DRAM.Source
- AMZNneutralmid-termThe host relays Jassy's view that Amazon allocates scarce memory by customer focus, but states no directional view on the stock.Source
- AI Capexneutrallong-termThe host notes new capacity can hit costs before revenue catches up, and management says margins may compress when those tranches land.Source
- AI Capexneutrallong-termThe host argues Anthropic's IPO story hinges on whether demand keeps outrunning a locked-in compute shelf, since revenue is mostly pay-as-you-go while the bill is mostly fixed.Source
- NBISneutralThe host says most neoclouds merely sell capacity, while NBIS is watching who is consuming that inference capacity; the company is in an AI inference demand wave with a real growth rate rather than a slogan, but no price direction is given.Source
- AI CapexneutralThe host asks that if personal AI agents are the wave Jensen also flagged, who captures the inference when it compounds; this is an observation on AI inference and compute demand with no explicit direction.Source
3 videos
- AI Capexpositive for stockslong-termCiting Jassy's figures, the host notes the generative AI business is growing several hundred times faster than early AWS, signaling unprecedented AI demand scale and pace.Source
- AMZNbullishlong-termThe host argues the market prices AWS as a mature cloud while its AI business reached a $25B run rate in 3 years — hundreds of times faster than early AWS — implying a mispriced second founding inside Amazon.Source
- Earningsneutralshort-termThis earnings season will see many beats yet red markets; it is a headline-driven market, so trading weekly options on reports is futile.Source
- AVGOneutralshort-termBroadcom is the very competitor the market fears for NVIDIA, yet it fell alongside it — an internally inconsistent move in the host's view.Source
- NVDAneutralshort-termThe host says NVIDIA falling on competition fears while the whole group is dragged down never made sense to him.Source
- AMDbullishlong-termAMD is still up 158% year to date because its business is expected to accelerate, though the business needs to catch up to the valuation; its data center growth and free cash flow are stronger than Intel's.Source
- RBRKbearishmid-termThe host is thinking of selling Rubrik completely because the valuation makes no sense when comparable companies are much cheaper, and the stock risks a 20–30% single-day crash on earnings.Source
- AI Capexpositive for stockslong-termThe host argues the AI trade is not over: every company says it is growing as fast as it can and is supply-constrained; the selloff is valuation catching up to price, creating opportunities.Source
- Geopoliticsnegative for stocksmid-termMidterm elections and Iran are near-term pressures on the stock market, along with possible year-end tax-loss harvesting; midterm years are historically weaker.Source
- AMZNbullishlong-termAmazon is extremely cheap for this type of business: AWS accelerated for a sixth straight quarter with a $578B backlog and record operating margins; a buy-and-hold name despite the after-hours drop.Source
- METAbullishlong-termMeta is grouped with Amazon, Google and Microsoft as a sleep-well buy-and-hold; the host sees nothing wrong with the business itself.Source
- GOOGLbullishlong-termGoogle, like Amazon, Meta and Microsoft, is a buy-and-hold name you can sleep well owning; it will be bigger and better five to ten years out.Source
- MSFTbullishlong-termMicrosoft is grouped with the other mega-caps as a buy-and-hold name that will be a bigger, better company years down the line.Source
- RDDTneutralmid-termThe host discloses a small Reddit position but moves it to the sidelines over stalled user growth and Google traffic declines; he would add only on catalysts like an AI data licensing deal, and thinks buybacks should accelerate at ~$100.Source
- INTCneutrallong-termIntel ran up far beyond its small turnaround, Q4 guidance missed badly, and the company's future largely depends on a successful Foundry; free cash flow only inflects in fiscal 2028.Source
- Otherneutralmid-termThe host says the interesting story is how little capital Anthropic consumed, posing growth, margins, or compute commitments as what IPO buyers should obsess over, without stating a direction.Source
- Otherneutralmid-termThe host argues Anthropic's $80.7B GAAP loss is mostly a $72.4B fair-value accounting mark, not cash leaving the business, with the real operating loss only $693M.Source
We only capture views expressed in public videos, each linked to its verbatim quote. Inclusion is not endorsement and not investment advice.