Podcast trade ideas

Invest Like the Best

Zach Reitano - Building Ro and Why Healthcare Should Be 50% of GDP - [Invest Like the Best, EP.500]

Nov 11, 2026102h

Positions · 1

• 12:23RoLongReitano says Ro is the leading GLP-1 provider, 10x'd revenue to a $3B run rate with 30% fewer people, and is building a vertically integrated patient-centric care platform riding three tailwinds. — Zach Reitano–
  Why this trade

  “Roe is the leading provider of GLP-1s in the country. We're probably signing up two, three, four, five times as many patients every single day as our competition.” — Zach Reitano

  1. Claims Ro signs up multiple times more GLP-1 patients daily than competitors and grew from $250M to $3B run rate.
  2. Positions Ro as demand aggregator integrating directly with pharma and bypassing PBMs.
  3. Result is a bullish long view on Ro as the scaled GLP-1 distribution platform (private company).

Ideas · 3

• 14:55NVOLongReitano says GLP-1 inventors should capture more margin than those scaling access, and DTC pricing lets pharma keep gross profit while prices fall and volumes surge. — Zach Reitano+1.2% since episode
  Why this trade

  “I think the inventors of GLP-1s should earn more, should grab more margin than the people who scale access to that.” — Zach Reitano

  1. Claims pharma going direct collapses the list-price-to-net gap and removes PBMs/insurers.
  2. Novo is one of the two leading GLP-1 manufacturers Ro partners with and whose volume is driving Ro's growth.
  3. Resulting posture is constructive on the drug inventors as volumes outstrip price cuts.
• 7:50LLYLongReitano cites Lilly's earnings as proof that GLP-1 price cuts are more than offset by volume, calling the class the most popular drug of our lifetime with room to 50%+ penetration. — Zach Reitano+0.7% since episode
  Why this trade

  “If you look at Lilly's earnings, prices come down, volume far outstrips it.” — Zach Reitano

  1. Claims GLP-1 demand is huge and growing, with majority of growth in cash pay.
  2. Notes that at Lilly, 'prices come down, volume far outstrips it.'
  3. Implies the drug makers benefit even as unit prices fall.
• 24:27GOOGLLongWaymo / driverless cars → Reitano argues driverless cars are 10-20x safer than human drivers and that regulators impeding them should bear a 'body count'; Waymo is the leading scaled robotaxi operator. — Zach Reitano-0.1% since episode
  Why this trade

  “Driverless cars in pockets have proven to be 10, 20 times safer than a human driver.” — Zach Reitano

  1. Claims driverless tech is proven dramatically safer than human driving.
  2. Alphabet's Waymo is the most established US robotaxi operator deploying this technology.
  3. A bullish autonomous-driving view maps most directly to GOOGL via Waymo.

Invest Like the Best

Ramtin Naimi on the State of Venture Investing - [Invest Like the Best, EP.499]

Nov 6, 2026222h

Positions · 4

• 39:36SPCXLongAbstract put ~15% of its 2023 vintage into xAI's formation round, a direct conviction bet on the frontier AI lab. — Speaker 1+9.1% since episode
  Why this trade

  “We put, I think, like, 15% of our 2023 vintage fund into the formation round of x AI” — Speaker 1

  1. Naimi says Abstract invested ~15% of its 2023 fund in xAI's formation round.
  2. xAI is a frontier lab with massive compute and capital needs.
  3. The firm's outsized allocation signals high conviction in the private name.
• 39:36POLYMARKETLongNaimi highlights Polymarket's explosive growth as a true consumer business and says Abstract was a big investor in its Series B. — Speaker 1-0.7% since episode
  Why this trade

  “we were a big investor in the series b round of Polymarket” — Speaker 1

  1. Naimi says Abstract was a big investor in Polymarket's Series B round.
  2. He calls the prediction markets' growth insane despite not being an AI company.
  3. He expects it will be interesting to see how prediction markets play out.
• 39:30CrusoeLongAbstract made passive growth investments into Crusoe, tying into Naimi's bullish view on compute demand and neo-cloud capacity. — Speaker 1–
  Why this trade

  “we've also advanced some passive growth investments into companies like Crusoe” — Speaker 1

  1. Naimi lists Crusoe among Abstract's passive growth investments.
  2. He argues compute demand from frontier labs and neo-clouds keeps rising with no end in sight.
  3. Crusoe is a direct beneficiary of that compute scarcity.
• 26:40ConduitLongNaimi's firm seeded Conduit nine months ago in brain-computer interfaces and believes it could define a trillion-dollar category. — Speaker 1–
  Why this trade

  “We seeded a company, like, nine months ago called Conduit” — Speaker 1

  1. Naimi says Abstract seeded Conduit nine months ago before BCI was a hot category.
  2. Its models show 80% accuracy reading thoughts and it now has 30,000 hours of data.
  3. He says Conduit's latest oversubscribed round prices in a trillion-dollar outcome.

Ideas · 10

• 6:19ANTHROPICLongNaimi repeatedly cites Anthropic as a trillion-dollar private company with 85% gross margins and voracious capital needs, calling it core LP exposure. — Speaker 1-11.4% since episode
  Why this trade

  “Anthropic's were 85% in the first half of this year” — Speaker 1

  1. Naimi notes Anthropic's 85% gross margins and trillion-dollar valuation trajectory.
  2. Anthropic is named among the handful of private companies that drive LP liquidity and exposure.
  3. Being long the best private compounders is a viable way to top-quartile venture returns.
• 25:06OPENAILongNaimi groups OpenAI with Anthropic and SpaceX as the three trillion-dollar private companies that anchor LP exposure and justify premium valuations. — Speaker 1-4.9% since episode
  Why this trade

  “between SpaceX and Anthropic and OpenAI, you have three trillion dollar companies” — Speaker 1

  1. Naimi says SpaceX, Anthropic and OpenAI are now three trillion-dollar companies.
  2. These few names generate nearly all LP liquidity and justify the frothy market's pricing.
  3. Owning the best private companies at high prices can still secure top-quartile venture returns.
• 13:15SPCXLongNaimi says LPs without SpaceX exposure have gotten essentially no venture liquidity, making it a must-have private position. — Speaker 1+9.1% since episode
  Why this trade

  “if you're an LP that wasn't in a fund that had a lot of exposure to SpaceX, you haven't gotten any liquidity” — Speaker 1

  1. Naimi argues LP liquidity is concentrated in SpaceX, Anthropic, OpenAI and Databricks.
  2. LPs lacking SpaceX exposure have received no meaningful distributions.
  3. Access to SpaceX is therefore a key driver of venture fund performance.
• 33:05DatabricksLongHost cites Databricks as a simpler, more tangible large private compounder and a core LP liquidity name alongside SpaceX and OpenAI. — Speaker 0–
  Why this trade

  “Whereas Databricks, where I think you're a big shareholder, it's just so so much simpler to understand” — Speaker 0

  1. Host calls Databricks a fantastic fast-growing large company they're a big shareholder in.
  2. Naimi agrees backing obvious great companies is a safe way to get exposure.
  3. But he warns this consensus strategy may not stay an easy way to make money.
• 28:45EtchedLongNaimi cites Etched as the proactive bet whose Sequoia-blessed $10B valuation proved startups could outcompete chip incumbents. — Speaker 1–
  Why this trade

  “it wasn't until Sequoia blessed them with a $10,000,000,000 valuation” — Speaker 1

  1. Naimi says Gavin Uberti's Etched was dismissed until Sequoia gave it a $10B valuation.
  2. Afterwards every other chip startup raised at astronomical valuations.
  3. Early, non-consensus bets on great founders are how to generate alpha.
• 44:18Isomorphic LabsLongNaimi ranks Isomorphic among next-generation AI-native biotech companies with $500B-$1T outcome potential. — Speaker 1–
  Why this trade

  “Isomorphic really high up on there in terms of IO and in terms of can you build AI native next gen Genentech” — Speaker 1

  1. Naimi names Isomorphic high among bellwether next-gen companies.
  2. He frames the opportunity as building an AI-native next-gen Genentech.
  3. Its market opportunity is $500B to $1T if it executes.
• 45:05Physical IntelligenceLongNaimi calls Physical Intelligence potentially a multi-hundred-billion to trillion-dollar robotics company with exceptional talent retention. — Speaker 1–
  Why this trade

  “I think physical intelligence is super interesting. I think that could be a multi $100,000,000,000, trillion dollar company” — Speaker 1

  1. Naimi says Physical Intelligence is super interesting with an exceptional team.
  2. He believes it could be a multi-hundred-billion or trillion-dollar company.
  3. Strong talent density and retention are what make him extra bullish.
• 44:18New LimitLongNaimi places New Limit high among next-generation biotech bellwethers with $500B-$1T outcome potential. — Speaker 1–
  Why this trade

  “I would put, like, New Limit and Chai Discovery really high up on there” — Speaker 1

  1. Naimi ranks New Limit high on his list of bellwether companies.
  2. He frames opportunities as AI-native next-gen Genentech-scale businesses.
  3. The team and access to capital support a $500B-$1T outcome.
• 44:18Chai DiscoveryLongNaimi cites Chai Discovery as a top next-generation AI biotech company with trillion-dollar potential. — Speaker 1–
  Why this trade

  “New Limit and Chai Discovery really high up on there” — Speaker 1

  1. Naimi names Chai Discovery high among bellwether companies.
  2. He sees AI-native drug discovery as a $500B-$1T opportunity.
  3. Best teams and capital access support this bet.
• 44:47Applied ComputeLongNaimi says Applied Compute pioneered the idea of companies owning their own data and knowledge intelligence, making it a category leader to watch. — Speaker 1–
  Why this trade

  “the company that probably pioneered that as the leader in the category is applied compute” — Speaker 1

  1. Naimi calls owning your own data and knowledge intelligence a unique idea.
  2. He says Applied Compute pioneered it as the leader in the category.
  3. He wants to watch how it executes going forward.

Invest Like the Best

Martin Casado on Frontier Model Economics, AI's Slowdown, and the Cursor Deal - [Invest Like the Best, EP.498]

Nov 4, 2026270h

Ideas · 2

• 36:09SNOWLongCasado argues AI will not disrupt durable data-infrastructure franchises like Snowflake; more software and more infra demand make it a good business that markets will want exposure to. — Speaker 1-0.6% since episode
  Why this trade

  “let's say that Snowflake is growing 30 to 40% and the economics are good... I don't think that they're at any risk of being disrupted by AI” — Speaker 1

  1. Casado says AI doesn't threaten infra like databases because operating the long tail of features/performance is very hard.
  2. He names Snowflake as a great example growing 30-40% with good economics, not at risk from AI (maybe Databricks).
  3. Resulting view: own Snowflake-type infra as a non-disrupted compounder.
• 20:54ORCLAvoidCasado says mature OLTP/transaction-database incumbents like Oracle face new AI-native database entrants with better properties that agents can use without retraining programmers. — Speaker 1-0.8% since episode
  Why this trade

  “OLTP is a transaction database. I think the Oracles of the world. That's a pretty mature industry” — Speaker 1

  1. He calls OLTP a mature industry with little movement, naming 'the Oracles of the world'.
  2. New AI-native databases can change their language because agents already know how to use them.
  3. Implies incumbent transaction-database positioning may be rebuilt over five years.

Invest Like the Best

Cliff Sosin - Owning the Complements to Intelligence - [Invest Like the Best, EP.497]

Oct 27, 2026463h

Positions · 3

• 55:40CVNALongSosin says Carvana's infrastructure and brand are so differentiated that few scenarios go really badly, justifying ~80% of his portfolio in it. — Cliff Sosin+27.0% since episode
  Why this trade

  “Carvana's infrastructure and brand and everything else is just so differentiated that there are few scenarios where it works out really badly” — Cliff Sosin

  1. Sosin states he is comfortable with Carvana at 80% of the portfolio because it is 'so differentiated' with few bad scenarios.
  2. He compares it to finding Walmart or Home Depot early, where concentration would have been correct.
  3. Resulting trade: hold an outsized long position in Carvana.
• 10:27METALongMeta owns genuine complements to intelligence — massive data and attention — so falling intelligence costs cut its spend while improving content, ads and targeting. — Cliff Sosin+0.3% since episode
  Why this trade

  “All they buy is intelligence. And if you think forward with the fact that the price of intelligence is falling by a factor of a 100” — Cliff Sosin

  1. Sosin argues Meta effectively spends its entire ~$250B budget on intelligence (data centers, inference, engineers, video).
  2. If intelligence prices collapse, its costs fall while its data/attention assets are not destroyed.
  3. Resulting trade: own Meta as a complement-to-intelligence beneficiary.
• 42:36HGVLongNamed as one of the roughly five holdings in Sosin's concentrated portfolio, alongside Carvana (80%) and Meta. — Speaker 0–
  Why this trade

  “It's 80% Carvana, vacations, 5% meta is basically your portfolio.” — Speaker 0

  1. Host describes Sosin's portfolio as ~80% Carvana plus Meta, Hilton Grand Vacations and a few others.
  2. Sosin does not dispute the composition, confirming it as a core holding.
  3. Resulting trade: hold HGV as part of the concentrated long book.

Ideas · 3

• 12:48GRNDLongSosin says Grindr has lots of attention and data but is constrained by a small team and budget, so cheap intelligence could let it do far more — though he does not own it. — Cliff Sosin–
  Why this trade

  “I don't actually own any Grindr... here, this mental model is what if you had 10,000 of Google's best engineers” — Cliff Sosin

  1. Sosin argues Grindr is limited by finite engineering/budget capacity rather than demand.
  2. Cheap ubiquitous intelligence could unlock the products a 10,000-engineer team could build.
  3. Resulting trade: a theoretical beneficiary, explicitly not owned.
• 28:34NVDAAvoidSosin notes the potential for AI to commoditize even Nvidia, since its moat is essentially a lot of people thinking hard — i.e., crystallized intelligence. — Cliff Sosin-25.6% since episode
  Why this trade

  “the potential for AI to commoditize even things like NVIDIA... NVIDIA quite literally makes flat files that they send to Taiwan” — Cliff Sosin

  1. Sosin says Nvidia 'quite literally makes flat files' and its edge is people who think really hard.
  2. In theory those could be replicated as intelligence gets cheap, and stepping back from the frontier lowers costs.
  3. Resulting trade: skeptical of the supply-side AI build-out, though not an active short.
• 29:26TSMLongSosin says TSMC's physical, hard-to-replicate process advantage has better attributes for enduring economics than pure intelligence, though copy attempts are a negative. — Cliff Sosin-7.5% since episode
  Why this trade

  “certainly, what they do has better attributes, I think, that regard... super hard to replicate and much less amenable to AI just because it is so physical” — Cliff Sosin

  1. TSMC built a super-hard, physical process moat over many years that is 'much less amenable to AI'.
  2. Physical scale and proximity are complements that persist even if intelligence commoditizes.
  3. Resulting trade: constructive but non-committal on TSMC as a supply-side holding.

Invest Like the Best

Christina Cacioppo - Building an Enduring Software Company in the Age of AI - [Invest Like the Best, EP.496]

Oct 20, 2026631h

Ideas · 8

• 10:00ANTHROPICLongCacioppo says token-based pricing pits apps against the models and that 'Claude token' comparison is a losing position, implying the model providers capture that markup rather than resellers. — Christina Cacioppo-30.6% since episode
  Why this trade

  “You've just marked up a CLOUD token forex. Are you forex as good as CLOUD?” — Christina Cacioppo

  1. Speaker claims anyone marking up a Claude token will lose the comparison because users can benchmark directly.
  2. That framing puts pricing power with the frontier labs like Anthropic rather than the application layer.
  3. Derived read-through: labs retain inference economics; a modest positive thematic view, not an explicit buy.
• 35:33OPENAILongShe cites OpenAI's RSA talk on models chaining benign issues into real exploits as evidence frontier lab capabilities keep advancing, supporting continued demand for their models. — Christina Cacioppo-14.8% since episode
  Why this trade

  “If you watch the OpenAI talk at RSA about hugging face, it's basically that.” — Christina Cacioppo

  1. Speaker references the OpenAI talk at RSA about hugging face as an example of model persistence finding chained vulnerabilities.
  2. That demonstrates frontier model capability gains that drive enterprise adoption of these labs.
  3. Derived thematic positive; no explicit buy stated.
• 36:48CRWDLongOn automating penetration-testing and remediation, she concedes 'CrowdStrike will do it better for an enterprise,' acknowledging its dominance in large-enterprise security. — Christina Cacioppo+0.2% since episode
  Why this trade

  “I think CrowdStrike will do it better for an enterprise.” — Christina Cacioppo

  1. Speaker says an agent-based ongoing pen-test/remediation vision is the direction for parts of Vanta.
  2. She explicitly says CrowdStrike will do it better for enterprise customers, where segment choice matters.
  3. Derived read-through: incumbent retains the upmarket; low-conviction positive.
• 31:46PLTRLongShe names Palantir as the company that 'does as well' with forward-deployed engineers because it has eight-figure contracts that make the model economics work. — Christina Cacioppo-15.8% since episode
  Why this trade

  “The company that we all think does as well is Palantir, which I think they probably do. They have 8 figure contracts.” — Christina Cacioppo

  1. Speaker critiques FDEs as unsustainable at five/six-figure contract sizes.
  2. She singles out Palantir as the exception that does well with eight-figure contracts.
  3. Derived positive only; not an explicit recommendation.
• 45:44Open-source model ecosystem (Llama/DeepSeek etc.)LongShe argues the certainty around open-source model providers is disconnected from the barely-built ecosystem around them, suggesting opportunity in tooling built on open-source models. — Christina Cacioppo–
  Why this trade

  “The ecosystem around them is so much less built out than the big labs. And that just feels off to me.” — Christina Cacioppo

  1. Speaker says discourse assumes open-source models will be very real, but their ecosystem is far less built out than the big labs.
  2. She says lab playbooks should apply to open-source models and there is more to do there.
  3. Derived thematic opportunity in open-source model tooling; not a listed call.
• 47:25RoboticsLongShe calls robotics 'underplayed' and underweighted versus software, adding that models are getting very good at it. — Christina Cacioppo–
  Why this trade

  “Robotics is probably underplayed, somewhat hot and hype y... it's still definitely underweighted compared to all the software stuff.” — Christina Cacioppo

  1. Speaker lists robotics as underplayed relative to all the software hype.
  2. She notes model quality for robotics is improving rapidly.
  3. Derived thematic long; not an explicit instrument call.
• 29:04Custom post-trained models (theme)AvoidShe calls custom training models for marketing value overhyped, saying inference providers admit there's no quality reason to post-train, only cost or marketing. — Christina Cacioppo–
  Why this trade

  “Custom training models for the marketing value... there is not a quality reason to do it.” — Christina Cacioppo

  1. Speaker lists 'custom training models for the marketing value' as overhyped.
  2. She reports inference providers say the only real reason to post-train is cost, otherwise it's marketing to investors.
  3. Derived avoid on companies pitching bespoke model training as a differentiator.
• 28:53Token-based pricing (theme)AvoidShe says token-based pricing is overhyped and that marking up model tokens invites direct comparison to the labs, a losing competitive position. — Christina Cacioppo–
  Why this trade

  “Token based pricing. Low margin for the sake of having low margins.” — Christina Cacioppo

  1. Speaker lists token based pricing among overhyped trends.
  2. She explains users benchmark token markups against Claude etc., so resellers lose.
  3. Derived avoid on businesses whose monetization is token resale margins.

Invest Like the Best

David Vélez - Building Nubank and the Future of Global Banking - [Invest Like the Best, EP.495]

Oct 13, 2026799h

Positions · 1

• 27:19BitcoinLongVélez says he's always been a Bitcoin fan and it plays a role in a diversified investment portfolio, especially as fiat currencies consolidate and people bank offshore. — David Vélez–
  Why this trade

  “And I've always been a Bitcoin fan, it plays a role in a diversified investment portfolio.” — David Vélez

  1. Vélez claims 190 fiat currencies could shrink to ~10 and people increasingly bank offshore in stablecoins and Bitcoin.
  2. That structural demand supports Bitcoin as a savings/diversification asset.
  3. He states Bitcoin plays a role in a diversified portfolio.

Ideas · 2

• 38:51US Incumbent Banks (large-cap banks)AvoidVélez argues incumbent US banks' high ROE comes from paying ~0 yield on $15T of deposits and charging fees, a 'margin is my opportunity' setup that digital banks will undercut. — David Vélez–
  Why this trade

  “That means your margin is my opportunity.” — David Vélez

  1. He claims banks pay ~40bps on $15T of deposits while rates are ~4.5-5%, a ~$600B annual cost to consumers.
  2. US incumbents have high ROE built on branch/fee structures vulnerable to digital undercutting.
  3. He frames that profitability as an opportunity for disruptors to take share.
• 26:32US Dollar / StablecoinsLongVélez believes US-dollar stablecoins democratize dollar access and will help the dollar stay the reserve currency, with billions banking on the offshore system. — David Vélez–
  Why this trade

  “US stablecoins ultimately will be the big savior or one of the biggest saviors of the US dollar because it democratizes access to the US dollar.” — David Vélez

  1. He claims stablecoins let billions in weak-fiat countries save in dollars.
  2. Dollar stablecoin demand is high while demand for weak local stablecoins is minimal.
  3. That supports dollar dominance though alongside 5-6 other fiat and crypto tokens.

Invest Like the Best

Andrew Huberman - The Frontier of Neurotechnology - [Invest Like the Best, EP.494]

Oct 6, 2026967h

This episode is largely a neuroscience and health-optimization conversation rather than an investing one, but Andrew Huberman makes one notable forward-looking prediction: Meta, OpenAI, Anthropic and Neuralink will ultimately become 'biotech companies' working on non-invasive read/write to the brain, and he sees sleep neurotech (an MIT-linked eye mask) as the first commercial wave. He explicitly says he wishes he had a stake in that sleep mask company but has none, and offers no stock-specific buy/sell calls; the actionable takeaways are thematic rather than direct trades.

Ideas · 6

• 24:23METALongNeurotech/AI-brain-interface theme → Huberman predicts Meta and the other big AI labs will ultimately be biotech companies racing to read and write to the brain non-invasively, opening a new frontier beyond LLMs. — Speaker 1-14.9% since episode
  Why this trade

  “Anyone that thinks that those are LLM AI companies is perhaps going to be surprised... Every single one of those is going to be a biotech company.” — Speaker 1

  1. Huberman says 'Every single one of those is going to be a biotech company' referring to Meta, OpenAI, Anthropic and Neuralink.
  2. Meta's large research budgets and Reality Labs hardware make it a listed vehicle for that neurotech/BCI race.
  3. Long META as an optionality play on its long-term neurotech ambitions.
• 24:23OPENAILongHuberman lists OpenAI among the AI labs he expects to become brain-focused biotech companies, with the stated goal of reading and writing from the brain non-invasively. — Speaker 1-21.9% since episode
  Why this trade

  “anyone who thinks that Meta, OpenAI, Anthropic, or Elon's got Neuralink as his stated neural interest ... is going to be a biotech company.” — Speaker 1

  1. Huberman claims the AI labs 'are all interested in the brain' and will pivot toward non-invasive neural read/write.
  2. OpenAI is private, so this is an unactionable thesis on the company itself.
  3. Direction is long on the theme, but no tradable instrument exists directly.
• 24:23ANTHROPICLongHuberman groups Anthropic with Meta and OpenAI as AI labs that will become neurotech/biotech companies pursuing brain read/write. — Speaker 1-34.3% since episode
  Why this trade

  “Every single one of those is going to be a biotech company. They're all interested in the brain.” — Speaker 1

  1. Huberman names Anthropic among companies whose 'real arms race of AI' is brain read/write.
  2. Anthropic is private with no direct listed equity proxy from the transcript.
  3. Theme-level long view only; not an actionable trade.
• 31:54NEURALINKLongHuberman sees Neuralink and its leader as part of the elite group driving non-invasive neural read/write and expects implantable neurotech to be the next big consumer trend. — Speaker 1-4.6% since episode
  Why this trade

  “I'm thinking implantable devices is where it's gonna go next.” — Speaker 1

  1. Huberman says implantable devices are 'where it's gonna go next' and cites Neuralink's neurosurgery head as an example of biohacking.
  2. Neuralink is private; no direct listed security identified in the transcript.
  3. Positive thematic view, not an actionable listed trade.
• 14:02Private sleep-neurotech eye-mask startup (MIT-linked)LongHuberman tried the MIT-linked eye mask that induces slow eye movements to speed sleep onset and amplify REM, says it works very well, and wishes he owned a stake; launch is 7-12 months out. — Speaker 1–
  Why this trade

  “I wish I had a stake in this company, I don't. These technologies are about seven to twelve months from release.” — Speaker 1

  1. Huberman claims the mask 'works very well' and helps amplify REM sleep.
  2. He explicitly states 'I wish I had a stake in this company, I don't,' confirming he has no position.
  3. Bullish on the product/company but not on any listed instrument.
• 26:25Eight SleepLongHuberman cites his Eight Sleep bed dynamically changing temperature to boost REM/deep sleep as proof that closed-loop sleep modulation is already working, a template for future neurotech. — Speaker 1–
  Why this trade

  “my eight sleep changes the temperature throughout the night to give me more REM sleep... It's already doing that.” — Speaker 1

  1. Huberman says his Eight Sleep 'changes the temperature throughout the night to give me more REM sleep' and it's 'already doing that.'
  2. Eight Sleep is private with no listed proxy in the transcript.
  3. Positive product validation as a precursor to direct neural write technology.

Invest Like the Best

Noah Shinn - Building Instinct: The Personal Agent - [Invest Like the Best, EP.493]

Sep 28, 20261154h

Noah Shinn describes Instinct, his invite-only personal AI agent that runs through messaging and voice channels, already driving over $1 billion in annualized transaction volume (half in travel) with ~10% day-over-day growth, and outlines a merchant-funded take-rate model rather than subscriptions or ads. The actionable thread is a long-term demand signal for AI compute and inference infrastructure, plus a claim that friction-reducing agents will lift transaction volumes at travel, ride-share and delivery platforms. No specific security is recommended or traded in the episode.

Ideas · 3

• 53:19NVDALongInstinct → Shinn says agents like Instinct will need compute 'orders of magnitude more than what we thought that we needed,' implying sustained accelerator demand across the AI buildout. — Noah Shinn-29.8% since episode
  Why this trade

  “I think the amount of compute that is going to be needed is going to be orders of magnitude more than what we thought that we needed.” — Noah Shinn

  1. Shinn argues proactive agents consume vastly more tokens and compute than earlier prompt-based copilots.
  2. NVIDIA is the dominant supplier of AI training and inference accelerators, so industry-wide compute growth flows to its revenue.
  3. Resulting view: long exposure to AI compute demand, though no one states a trade.
• 16:05InstinctLongShinn says Instinct is growing 10% day over day, has over $1B annualized transaction volume, and plans a merchant take rate, framing it as a potentially huge consumer platform. — Noah Shinn–
  Why this trade

  “we're approaching over a billion dollars a year in transaction volume through the platform” — Noah Shinn

  1. He claims the product grew purely by word of mouth with $0 marketing and 80% retention once sensitive data is connected.
  2. He plans to monetize via merchant distribution fees similar to Apple Pay, Amex or Shopify, targeting travel's high take rates.
  3. Resulting view: positive on the private company, no listed security available.
• 35:12UBERLongShinn argues frictionless, proactive agents could raise ride-share and delivery volumes, saying reduced friction means 'that 30% looking a lot bigger' for underlying-service businesses. — Noah Shinn-0.7% since episode
  Why this trade

  “What would that do to Uber's business or any rideshare business if now the default for existing riders that love Uber” — Noah Shinn

  1. He claims Instinct can pre-book cars based on calendar, cutting friction for existing riders.
  2. Uber is named as an example where more transaction volume at less app attention benefits the business.
  3. Resulting view: structurally positive, but he explicitly says he doesn't know Uber's numbers.

Invest Like the Best

Gabe Stengel - Building Investing Superintelligence - [Invest Like the Best, EP.492]

Sep 22, 20261303h

Ideas · 5

• 18:32RogoLongGabe argues Rogo can be the systems-of-record and exchange layer for private-market dealmaking, a niche the AI labs won't build and where incumbents lack an AI-native harness. — Gabe Stengel–
  Why this trade

  “we can get to $5,000,000,000 in revenue by going deep across those things and creating the systems of record” — Gabe Stengel

  1. Gabe says finance has huge, messy, vertical-specific TAM that general model labs will not address.
  2. Rogo is an applied-AI platform wiring model intelligence into deal-maker workflows and data rooms.
  3. He expects a multi-billion-dollar durable franchise from going deep across these niches.
• 33:33JPMLongGabe cites JPMorgan's announced expansion into SMB M&A as an example of an incumbent using AI-driven productivity to enter markets previously uneconomic to serve. — Gabe Stengel-2.3% since episode
  Why this trade

  “JPMorgan just announced that they're gonna try and do a lot more m and a work for SMBs.” — Gabe Stengel

  1. Gabe claims JPM announced it will do more M&A work for SMBs using AI-enabled one-person deal teams.
  2. That implies JPM can capture new fee pools it previously couldn't staff economically.
  3. The proxy trade is JPM gaining share in underserved advisory segments.
• 19:42ANTHROPICLongGabe repeatedly points to Claude models and Claude Code's superior harness as driving usage and token consumption, arguing harness design unlocks model capability and revenue. — Gabe Stengel-36.5% since episode
  Why this trade

  “the harness and the way that it was presented from Claude was far better and it just allowed the models to exercise more of their long running capabilities” — Gabe Stengel

  1. Gabe says Claude Code's harness let models exercise long-running capabilities, causing a usage run-up.
  2. He notes Anthropic's token-consumption pricing makes its revenue scale rapidly with customer usage.
  3. The implied view is durable growth in Anthropic's token-based revenue.
• 18:32OPENAILongGabe frames OpenAI as one of the frontier labs that will capture enormous token-consumption revenue across enterprise software and analytics. — Gabe Stengel-21.7% since episode
  Why this trade

  “you're on the pathway of trying to go from a 100,000,000,000 in revenue to a trillion in revenue” — Gabe Stengel

  1. Gabe describes labs moving from $100B to $1T revenue via token consumption.
  2. He contrasts token brokers like Cursor and Claude Code versus per-seat enterprise sales.
  3. OpenAI is positioned as a beneficiary of that abundant token demand.
• 26:28CursorLongGabe lists Cursor among token-consumption 'token broker' businesses that can grow commercially with far fewer people and elevated API usage. — Gabe Stengel–
  Why this trade

  “there are the token brokers, the token consumption businesses where you're selling in the parts of enterprise where they're used to buying usage based tools. Cursor, factory, cloud code and others.” — Gabe Stengel

  1. Gabe categorizes Cursor as a token broker selling in usage-based enterprise segments.
  2. Such businesses can scale with fewer humans than per-seat enterprise sales.
  3. Implied view: Cursor benefits from abundant token demand.

Invest Like the Best

Michael Moritz - Lessons From 40 Years of Investing and Writing - [Invest Like the Best, EP.491]

Sep 16, 20261447h

This is a wide-ranging conversation with Sequoia veteran Michael Moritz about career, family history, writing and leadership rather than a stock-picking episode. The only market-relevant views are his praise of Elon Musk as possibly the greatest American industrialist and his broader optimism that AI will create more jobs than it destroys; there are no explicit buy, short or avoid recommendations on listed securities.

Ideas · 2

• 54:37TSLALongElon Musk ventures → Moritz calls Musk possibly the greatest industrialist in American history and is 'gobsmacked' by his achievements across industries, an admiring founder view rather than a valuation call. — Speaker 2+7.9% since episode
  Why this trade

  “He may well be the greatest industrialist in the history of America.” — Speaker 2

  1. Moritz says Musk may be the greatest industrialist in US history and is gobsmacked by his multi-industry achievements.
  2. Tesla is Musk's flagship public company and the primary listed vehicle for his operating track record.
  3. The implied stance is favorable on Musk-led innovation, not an explicit buy recommendation.
• 1:00:17AI infrastructure / broad marketLongMoritz is more hopeful than fearful on AI, arguing it will create more jobs than it destroys and that prior technology waves ultimately raised long-term employment. — Speaker 2–
  Why this trade

  “I think that it's gonna create more jobs than are gonna be lost.” — Speaker 2

  1. He notes nuclear and genetic Armageddon fears never materialized and expects AI to create more jobs than it destroys.
  2. That implies durable long-term economic benefit rather than a specific tradable instrument.
  3. No actionable listed security is named, so this is a thematic, non-explicit view.