Sam Andrew (@samuelmandrew) header
Sam Andrew (@samuelmandrew) profile picture

@NorthIslandVC | Former hedge fund analyst (BlueMountain Capital) | Not financial advice

New York, NY
Joined October 2011
Photo by Sam Andrew (@samuelmandrew)Photo by Sam Andrew (@samuelmandrew)Photo by Sam Andrew (@samuelmandrew)Photo by Sam Andrew (@samuelmandrew)Photo by Sam Andrew (@samuelmandrew)Photo by Sam Andrew (@samuelmandrew)Photo by Sam Andrew (@samuelmandrew)Photo by Sam Andrew (@samuelmandrew)Photo by Sam Andrew (@samuelmandrew)Photo by Sam Andrew (@samuelmandrew)
Meta just entered AI with Muse. It's a foundational product pioneering a new category. And a potential game changer for $META - 4 Muse predictions: 1/ Muse is the first example of Meta developing a great AI product. That goes a long way to change the narrative on Meta wasting AI capex $. 2/ It's a signal that Meta can ship AI products that create a whole new category, which is additive to their business and doesn't compete with leading leading labs products. It's mass market AI. 3/ Muse will be a cost sink for a while. Muse is probably loss making per user. But land grab is the most important thing now, improved unit economics will follow after. 4/ Muse growth will be constrained by Meta's compute capacity. Zuck will have to make the hard call of whether to allocate his scarce compute to Muse or MSL.
Introducing Muse, the personal agent that understands your goals and works 24/7 to get things done for you.
2
274
Q2 results just out showed the acceleration of the Figure Connect business. A resounding endorsement of their pivot to an asset light business $FIGR
$FIGR: The Market May Be Underwriting The Wrong Business. The market still sees a lender. But 78% of volume is now partner-originated, Connect is scaling rapidly, and the highest-margin fee lines are becoming a much bigger piece of revenue. We think Figure is becoming something else entirely. Our new deep dive with @NorthIslandVC:
3
2
11
1,078
Some valid points
Since i'm a solutions guy, and not a problem guy. Here is what I would suggest for $GLXY This should not be a hard story to tell. First, stop overpromising and underdelivering timelines for leases (or admit that you had a deal, and it fell through, which is the only theory that makes any logical sense). A deal falling through isn't a bad thing as the market is re-rating higher anyway. But loss of confidence because you missed a deadline and then failed to address it is a problem. Second, stop hiding from the crypto volatility and embrace it. Blockchain (along with AI and robotics) are the two hottest technologies right now. People love blockchain right now but can’t figure out how to invest in it. They don’t want bitcoin:native anymore.  Layer 1s are a joke. They want exposure to the only 3 areas of crypto that are actually growing: 1) RWA tokenization 2) stablecoins / payments 3) Defi GLXY gives you exposure to all of three of those. So in conclusion, pitch GLXY as "the stability of an AI data center with the upside volatility from blockchain's hottest areas of growth" GLXY should be the hottest stock on the planet. Gives you exposure to all 3 of those + AI Just need to tell the actual story and stop bouncing around like you ran out of lemons at the lemonade stand and are now selling sugar water.
1
3
202
A great pod as always. A few notable things to me: 1. Sam's not concerned about China distillation bc of inference margins -> that's bullish for the AI capex ROIC debate 2. OpenAI still doesn't have enough compute -> bullish for co's in compute stack 3. He thinks competitive advantages are in compute not intelligence -> again bullish for co's in compute stack 4. AI robotics will have their ChatGPT moment in 2-4 years -> lookout for infrastructure implications 5. AGI has not upended economy as ppl thought it would -> surprising to hear, may a way to dispel growing societal AI angst
My conversation with Sam Altman (@sama), CEO of OpenAI. We discuss: - Kimi, distillation, and open source - OpenAI's compute bets - The Hugging Face incident - What happens after AGI - Raising kids in an age of abundant intelligence - And much more Enjoy! TIMESTAMPS 0:00 Intro 4:10 The Race for Compute 14:24 A Sci-Fi Cyber Incident 16:14 The Promise and Risks of AGI 23:27 How AI Will Change Jobs 29:38 Sam’s Vision for a Personal AI 35:02 Robotics, ChatGPT, and What Comes Next 44:39 The Weight of Leading OpenAI 51:36 OpenAI’s Biggest Lessons
1
168
The archetype of founder we look to back 🚀
Every so often I meet a founder who completely rewires how I perceive the future. It’s a rare feeling. But when I find that connection, I don’t forget it. This is my singular goal, finding those that recreate this feeling in not just me but everyone they meet. At NIV we call them Missionary Empire-Builders. These are the highest agency people among us. Their hardest life experiences forge real purpose in everything they do. They think for themselves instead of absorbing the crowd, even when there is personal consequences. And they execute relentlessly in a way that seems alien to most. They change the world by closing the gap between their conviction and the consensus. It’s been a great honor supporting these entrepreneurs in their conviction and mission at NIV: @bryan_johnson & @_katetolo (Immortals), @erickbrimen (Prospera), @ErikVoorhees (Venice), @nasdaily (Nas), @mcagney (Figure), among others. See below for our full thesis on what we think makes the founders we support at NIV so special: docsend.com/view/gaj55hu48v7…
1
3
208
What's so compelling about $HOOD ?
Robinhood may have copied Amazon's smartest idea. Robinhood Gold isn't just a subscription; it's a trojan horse to get users to move their entire financial life onto the platform. More assets = more ways to monetize. Trading fees are just the beginning. FT @BitcoinJesusETH @samuelmandrew Tune in to know more ⏱ TIME POINTS ⏱ 00:00 – Intro 01:32 – Why Robinhood Could Win 04:37 – PRO Is 33% Off 05:01 – Prediction Markets Explained 08:39 – How Big Can They Get? 11:16 – Robinhood's Growth Strategy 15:08 – Sponsor: Securitize 15:46 – Sponsor: Bitget 16:20 – Robinhood's Amazon Prime 19:22 – Robinhood vs. Coinbase 21:46 – Why Tokenization Matters 24:15 – Robinhood's 7% Yield 26:04 – Meta's Super App Threat 28:37 – What Breaks the Bull Case? 29:37 – What's Next for Sam?
230
Meta is evolving its AI business to be more like other hyperscalers. I don't think this is an indication of weakness in AI compute bc i) Meta suggested they could sell capacity for a higher price than they bought it, ii) its a shift in capacity mix, not net new, iii) Nvidia would be down if the market was really worried about $/GPU pricing. Meta is essentially becoming more like other hyperscalers - it'll use its capacity internally and sell it externally. The market has been rewarding that strategy bc it makes underwriting the AI capex ROI simpler.
BREAKING: Meta, $META, is developing a cloud infrastructure business that will sell access to AI compute and aims to compete with Amazon, Microsoft, and Google, per Bloomberg. Meta shares are surging over +7% on the news.
315
Robinhood is cooking
HOOD released a very truncated version of June monthly metrics, between June 1 - June 25. This covers 18 trading days for equities/options and 25 for crypto/PMs (90% and 83% of June, respectively). Extrapolating the numbers through month-end linearly, noting Q/Q numbers represent the entire Q2 period: • $381.1B of equity volume (+21% m/m, +48% q/q) • 304.4B of options volume (+32% m/m, +30% q/q) • $16.8B of crypto volume (+38% m/m, -38% q/q): • $7.2B Robinhood App (higher take rate) (+22% m/m, -22% q/q) • $9.6B Bitstamp (institutional) (+52% m/m, -47% q/q) • 6.3B prediction markets contracts (+62% m/m, +52% q/q) Note the PM extrapolation will likely come out higher than 6.3B. This is because the World Cup volumes only started to pick up in the second half of June. I wouldn't be surprised if we saw closer to 7B. These numbers are nothing short of remarkable. All metrics including crypto have shattered my expectations, and June's RH crypto and prediction volumes represent potential upside to both buyside and sellside expectations. While equity and options may have been priced into the stock I do not think crypto or PMs were.
266
Amazon Prime + generational wealth transfer + prediction markets -> all could be coming to $HOOD @NorthIslandVC and @artemis explain how
8
7
39
24,942
The next trillion-dollar customer doesn't have a pulse. AI agents are about to shop, invest, and transact on our behalf - and the entire financial system was built assuming a human on the other end. Money for Machines is our (@NorthIslandVC) deep dive on what replaces it and what needs to be built; read the whole thing 👇 docsend.com/view/yhr9tzs5mcp…
2
6
752
Wonder why NDX just had one of its wildest intraday moves in the last 3 years....levered ETF trading. SOXS 3x levered short semi ETF just had a trading day that rivals GFC
1
1
1
162
Money For Machines - our agentic finance thesis @NorthIslandVC
At NIV, we've been seeing tremendous growth clustering around the intersection of AI, Stablecoins, DeFi, and Fintech. This has led us to believe that agentic finance, defined broadly, will be one of the biggest trends of the next decade. Indeed, we believe that 1) AI agents are on the cusp of becoming true economic actors, able to discover, negotiate, and transact, 2) there will be billions of such agents within 5-10 years, engaging in trillions of dollars in annual transactions, 3) these agents will need new financial systems, markets, infrastructure, and tools, which will unlock applications and experiences that are hard to imagine today, and create tremendous value. This is one of our main areas of focus, so if you're working in this space, please get in touch. Full Deck Here: lnkd.in/emSGsGDK
2
161
Robinhood getting into the IPO business...a new business line?
Robinhood Securities is now approved to serve as an underwriter. Since IPO Access launched in 2021, we've watched retail go from an afterthought to a key part of how companies plan an IPO. The question changed from "why allocate to retail at all?" to "how big can the allocation be?" Becoming an underwriter, and not just a selling group member, is the natural next step to better serve our customers and our issuers. We intend to be disruptive in this space.
2
268
productivity gains anyone?
Chart of the Day: AI is bending the knowledge production curve across multiple domains.
1
1
117
Regulated US crypto perps 🚀 Could other perps be next?
In my first public remarks as @CFTC Chairman, I made clear that the agency would use the tools at its disposal to onshore crypto asset perpetuals. Today, the @CFTC delivered on that commitment. This morning, the @CFTC took historic action to permit the listing of a true bitcoin perpetual contract by a CFTC-registered exchange, charting a path for one of the most liquid segments of the crypto asset markets to exist within the US regulatory framework.
2
1
1
308
ICE CEO & Founder very constructive on Hyperliquid. He's publicly stating how impressed he is with the team, how novel what they're doing is, how they're collaborating, and how regulation has to change. He sees the SpaceX IPO (and the respective pre-market on Hyperliquid) as another upcoming test to see gage how relevant Hyperliquid is. Stay tuned!
Jeff Sprecher, founder and CEO of $ICE (owns the NYSE) on Hyperliquid: "This Hyperliquid that we're talking -- if you haven't heard about it, it's bigger than NASDAQ, okay? It's 11 people. You look at it, you're like, wow, that's pretty something." If it wasn't clear before, hyperliquid:native has grown far beyond crypto. The incumbents have noticed, are paying close attention, and even spending time with the team Bernstein excerpt below and worth the read imo:
1
198
This chart from GS encapsulates what's driving equity market moves.
71
Capitalism at its finest. Anthropic booming. TSMC preventing a bubble. Distorted valuations. Lots brewing in this great pod: 1. We're witnessing the most extreme moment in capitalism history. Anthropic recreated the 3 best SaaS companies in the span of a few months this year. 2. Anthropic's revenue should be even higher than people realize, maybe ~$200b. That's because they're purposely throttling due to compute shortage. Means Anthropic could actually be valued at 5x "unconstrained revenue." Anthropic is also the most efficient token usage model; use ~80% fewer tokens than OpenAI. 3. TSMC is single handedly preventing an AI bubble. They're limiting capacity expansions, which is so far preventing an overbuild that is common of every bubble. The likes of Intel and Samsung could force that discipline to break, which would be more concerning for a bubble. 4. Power shortage gets solved by 2027-2028. Racks in space is part of the answer. 5. Cross-sectional valuations are distorted. Stocks prices of poor commodity businesses are mooning; typical of commodity super cycle whereby the highest cost product benefits the most (amplified by retail). But higher quality names are trading at reasonable valuations.
This is my sixth conversation with @GavinSBaker. As always with Gavin, the conversation covers a lot of ground, but we spend the most time on watts and wafers. We discuss: - Why the wafer shortage may prevent an AI bubble - Data centers in space (reframed) - Elon's Terafab and the new chip companies challenging Nvidia - Usage-based pricing - The disaggregation of GPUs - DRAM, frontier tokens, and open source Enjoy! Timestamps: 0:00 Intro 7:55 Anthropic and OpenAI Valuations 12:58 Watts, Wafers, and Infrastructure 14:39 Orbital Compute and Data Centers in Space 22:49 Avoiding the AI Bubble 28:26 Terafab and the Future of US Manufacturing 32:16 Returns to the Frontier 37:23 Continual Learning 42:03 New Chip Companies 48:52 Extending GPU Lifespans and Private Credit 51:22 The Application Layer 57:32 The Token Path and Open-Source Dynamics 1:01:37 Cybersecurity 1:05:46 Diversity Breakdown 1:11:59 Assessing the Big Tech Players in AI 1:19:02 Geopolitics, Personal Safety, and the AI Horizon
1
124