BuildAndBet

Wild Prediction: While Anthropic is paving the path and leading the charge in enterprise AI, the long term winner will be $MSFT. They already own distribution and it's only a matter of time until they leverage OAI to build competing enterprise services. Distribution always wins.

@LD4Runner @KrisPatel99 @Kross_Roads You’re selling ATM calls?

@KrisPatel99 @Kross_Roads I sold my MSFT months ago, still holding 200+ NVDA and making min $500 a week selling cc’s. It’s hard to sell when your dropping your cost basis $2.50 a week.

$MSFT and $NVDA are both incredibly attractive here, but I'd pick Nvidia easily. Three reasons: 1⃣ Nvidia is at the epicenter of the AI wave, which looks to continue for a while (likely surpassing most estimates). Microsoft is positioned decently and not going away, but I'd rather stick with the prime beneficiary rather than worry about software's moat. Further, with Microsoft and most of the hyperscalers, there are significant CapEx concerns. These concerns mostly benefit Nvidia. Unless you believe China invades Taiwan, or AI is a passing fad, it's just a cleaner bet. 2⃣ While forward P/E looks similar over the next 12 months, when we look 2 years out, it's clear $NVDA is more attractive. 3⃣ FCF trumps all, and the P/FCF of Nvidia is lower. And on a forward basis, substantially lower than $MSFT. Both are, over the long term, great opportunities, but Nvidia's returns will likely dwarf that of Microsoft.

@Kross_Roads @amitisinvesting $MSFT is the right choice here IMO. The -21% discount creates a great margin of safety that $NVDA's -5% doesn't. NVDA also has a lot of sensitivity to China demand and any of the hyperscalers not blinking on CapEx, relative to MSFT. NVDA has a higher upside but also higher risk

Because AI is real and Nvidia is at the epicenter. Microsoft is a steal here too, but unless you believe in China invading Taiwan or AI being a passing fad, it's hard to pick them over Nvidia. P/E in 12 months looks similar. In 24 months? Nvidia crushes MSFT. And forward P/FCF? It's not even close.

Would you rather buy $MSFT or $NVDA at these levels for a one year holding period? $MSFT — $370, -21% YTD, $2.7T MC, 22x FWD PE $NVDA — $179, -5% YTD, $4.3T MC, 21x FWD PE

@eglyman @tryramp At what rate is the AI spend going up in proportion to the revenue growth?

Since 2023, the top quartile of AI spenders on @tryramp have more than doubled their revenue. Bottom quartile? Flat A roofing company in Texas. A window installer in Utah. A construction firm in Florida that grew 65% The gap is accelerating and most companies don't feel it yet

@jukan05 It started falling before Turboquant announcement though?

The fact that memory stocks are crashing because of Google’s Turboquant is a pretty good indicator of how many clueless people this market is filled with. It’s like saying Aramco should crash because Toyota came out with a next-generation hybrid engine.

@firstadopter People in early 2000s probably - "Internet companies still use a lot of Newspapers"

OpenAI and Anthropic still use a lot of SaaS software!

@1shankarsharma @sandeep_PT At this point any "deal" can be claimed as victory no?

@sandeep_PT Indeed. So America cannot Retreat as a loser from the war. So what option does it have?

Look, based on almost every assessment, the ground invasion will result in 1000s of American deaths. So why do it? Maybe that's exactly why? Soldiers die. Body bags on TV. Public anger. Calls of " Nuke 'em". Under the cover of public anger, they then do exactly that....

@KrisPatel99 Have you bought already or waiting?

The amount of fear being pumped right now is hilarious... Keep at it boyzzz Im looking for discounts =).

@TaylorHoliday @seyitaylor For items less than $20 probably

@seyitaylor Tik tok shops is changing everything in the US. The debate is over. Social commerce is real.

Uhhh is this the beginning of the inevitable breakup between Stripe and Shopify? This is a big deal. It’s also a bunch of my predictions coming to life at once. Here's what's happening: Stripe is powering a native checkout inside Facebook ads. You see an ad, tap "Buy now," and purchase without ever leaving Facebook. Powered by Stripe's new Agentic Commerce Protocol (ACP)*, an open standard built with OpenAI. This is not the breakup. It's worse. It's Stripe building the layer that makes Shopify optional. Here's the threat chain: 1. Checkout moves off the storefront.* If customers buy inside Meta ads via Stripe, they never visit the Shopify store. The store becomes a warehouse backend, not the point of sale. 2. Stripe becomes the merchant of record infrastructure.* ACP lets any AI agent (ChatGPT, Meta, whoever) trigger a purchase. The business keeps merchant of record status, but the commerce infrastructure is Stripe, not Shopify. 3. Shopify Payments IS Stripe under the hood. Shopify's payment processing is a white-labeled Stripe integration. Stripe going direct to Meta means they're routing around their own reseller. Why let Shopify clip the ticket when Stripe can own the relationship directly? 4. The "agentic" angle is the real play. ACP isn't just Meta. OpenAI is the first AI platform to implement it. This means ChatGPT, and eventually every AI assistant, can trigger purchases through Stripe without touching a storefront. That's a future where "browse a website and add to cart" becomes a legacy behavior. For brands specifically This changes the ad-to-purchase funnel fundamentally. If a Prophit Engineer is optimizing Meta ads and the purchase happens inside the ad unit via Stripe checkout, the attribution model changes, the role of the landing page changes, and creative becomes even MORE important because the ad IS the store. Brands should be watching this closely. Short-term, nothing changes. Medium-term (12-18 months), the brands that figure out how to sell inside AI surfaces and ad units via ACP will have a structural advantage. So is it the breakup? Not yet. But it's Stripe signaling that they don't need Shopify as the distribution layer anymore. Stripe + Meta + OpenAI is a commerce stack that doesn't require a storefront. Shopify's response will tell you everything: do they build their own ACP competitor, or do they lean into being "the backend" for agentic commerce?

@KiroIkigai @Midnight_Captl That is now. NVDA has hovered around $175 / $180 since the last 6 to 7 months I think. Also, shouldn’t risk off in fact help NVDA because that is a safe stock? Market knows CapEx isn’t slowing down and majority of it is going to them.

@BuildAndBet @Midnight_Captl The macro economic factors are all painting risk off

$META head of Infra is on stage now talking about how demand for infra is going exponential “to put it mildly” Yeah CapEx is not slowing down guys… If you think it is, just get the thought out of your head. We’re nowhere near peak.

@Midnight_Captl @KiroIkigai What explains $NVDA not catching a bid despite the crazy growth and CapEx numbers?

@KiroIkigai It’s not slowing down. The whole ecosystem is signaling it but people still don’t believe.

@realroseceline Love your analysis framework and write up. How much time did this analysis take? (Not the write up, just the analysis)

Thoughts on $ZETA Lots of you asked me about $ZETA, so I took a quick look at the financials and wanted to share a few early thoughts. This is actually an interesting one because it looks messy on the surface, but there’s something happening underneath. Sales are strong, revenue grew from $458m in 2021 to $1.3b in 2025, and it has been fairly consistent along the way. Gross profit was ~$790m on $1.3b of revenue, so roughly 60% margins, which means the core product is not the issue. The problem has been everything below that line on the income statement. SG&A is extremely high at ~$575m or about 45% of revenue, and it has not scaled well, which is why operating leverage has been nonexistent. R&D at ~$120m is fine on its own, but combined with SG&A it overwhelms the business. The key is that operating income just turned roughly break even in 2025 at ~$5m after years of losses. Operating expenses grew only about ~$100m while revenue jumped by ~$300m, which is the first sign that incremental dollars are starting to fall through. This is not a broken business, it’s a high growth company that historically lacked discipline and is now starting to mature, although non operating losses and still negative pretax income show it is not fully there yet. The balance sheet is not weak, but it is not strong either. Liquidity is fine with ~$320m in cash and ~$685m in current assets against ~$430m in current liabilities, and debt at ~$200m is manageable so there is no survival risk. What stands out is how much of the asset base is goodwill and intangibles, roughly ~$776m combined, which is more than half of total assets. That tells you growth has been acquisition driven to some extent, and if those deals do not deliver, you risk impairments and poor capital allocation. Retained earnings are deeply negative at around -$1.06b, meaning shareholders have funded the business historically, but equity is now growing quickly, which aligns with the move toward profitability and improving efficiency. Current liabilities jumped from ~$200m to ~$430m is something to watch. Some of it may be normal for growth or deferred revenue, but it can also mean working capital pressure. Then you get to the cash flow statement, which is where things look better than I expected after reviewing the income statement and balance sheet. Operating cash flow is ~$200m on $1.3b of revenue, about 15%, and it has grown consistently from ~$78m to ~$200m over a few years. But a big part of that comes from stock based compensation ~$178m, which means a large part of the cash flow is dilution. Even so, free cash flow at ~$185m with low capex shows this is an asset light model with strong economics. What is interesting is the capital allocation. After issuing a large amount of stock in 2024, $ZETA bought back around ~$121m in 2025. That suggests a transition from relying on shareholders to starting to return capital, which is a meaningful change if it continues. There are still risks. Receivables are a drag on cash, down ~$77m, which means they are waiting longer to get paid. Acquisitions are still ongoing, around ~$90m, which comes back to the goodwill and raises the question of how much growth is truly organic. So when you put it all together, the income statement says barely profitable, while the cash flow statement says this is already a decent but loosely managed business. The truth is probably somewhere in the middle. This is a company with strong underlying economics that converts revenue into cash, but still masks part of its costs through SBC and depends on acquisitions. If they keep growing operating cash flow, bring SBC down as a percentage of revenue, and grow organically, $ZETA can become a quality cash generator. If not, it will continue to look good on the surface while largely redistributing value through dilution and acquisitions. 🌹

@harleyf @Shopify @ChatGPTapp Experience is very buggy. How to report bugs?

Brands on @Shopify are now shoppable inside @ChatGPTapp. AI shopping isn’t coming. It’s here. As always, our merchants are best positioned. Let's go.

@GavinSBaker @WSJ I think your long response to that article is giving it more attention than the article itself.

Strange @WSJ article about Jensen renting out the San Francisco opera house last year for a performance of “The Monkey King.” The article essentially frames it as a gilded age “champagne and opera” style event for “AI’s kingmaker.” Makes for a good story, but the reality was quite different. Mostly because the article leaves out an important detail; Jensen’s stated reason for hosting the event: helping to revitalize San Francisco’s downtown and the opera house. This was the actual focus of his speech, not the “champagne and opera” joke. And the event was full of Jensen’s good friends as well as various AI figures. When I spoke with Jensen that night at the opera, the first thing he said to me was that it was really personally important to him to help downtown San Francisco get back to its glory days. It was essentially a fundraiser for the opera and San Francisco where no one was asked to contribute.  As someone who gets asked to contribute to a lot of charities and go to a lot of charitable events, I thought this was a generous and classy gesture on Jensen’s part and I’m sure many attendees took it upon themselves to donate. I certainly did! Article would read differently with these facts, which kinda seem like the most important ones to me. As far as the illustrious AI figures in attendance, important to note that the CEO of Jensen's largest accelerator competitor was actually at the opera at Jensen's invitation the night I was there; which again is quite at odds with what is described in the article. And please note that Nvidia’s largest competitor is not actually AMD. One of the defining features of tech in the 90s was that the CEOs generally hated each other. I think it is good for the world and the AI industry for someone like Jensen to bring the industry together and I'm happy that he is doing this. If anyone believes that any of the companies referenced would’ve struggled to raise money without Nvidia’s participation - they are simply wrong. I know because I was around many of these fundraises. And OpenAI’s embrace of AMD is object evidence that companies who have raised money from NVIDIA feel free to use competing products. Finally, kinda sad that the net effect of this article might be to make other CEOs less likely to host comparably civic minded events. I hope that I am wrong and that both Jensen - and other CEOs - are undeterred by this article. Nice for people to give back.

@GergelyOrosz Tbh applying for a job via a link has been dead for a long time.

The death of inbound applications is upon us: and yes, it’s in a big part because of AI making it dead simple to apply. And so inbound applications become noisy, with increasingly more of non-qualified people. And so companies rely on referrals and recruiters to source instead.

Taste as a moat is overhyped IMO: 1. Optimizing for "better" leads to diminishing returns 2. Eventually the user just ships the 'Nth' output without caring much about the taste 3. Users now tolerate average designs as far as the real value from the tool is clear.

@amitisinvesting Answer would've been the same if the question said "$6 trillion"

$NVDA “Is it possible for NVIDIA to be a, you know, $3 trillion revenues company in the near future? The answer is of course yes. And the reason for that is because it’s not limited by any physical limits. There’s nothing that I see that says, you know, gosh, $3 trillion is not possible."

Nuclear for powering AI ❌ AI for Nuclear collaboration ✔️

@illscience Which companies get impacted?

if you thought saas-pocalypse was bad just wait for computer use to get really good later this year the implications for incumbents are 100x more than coding agents because computer use asymmetrically benefits “hostile” integrators & expect a race to commoditize complements

@deedydas Which machine are you using? Any lag introduced over time?

Uses Whisper TTS for speech recognition, Qwen 14B as the LLM and Kokoro for voice.

Siri has been broken for 13 years so I built my own. Completely on-device. No internet needed. Controls my Mac, sets reminders, fetches live data, answers questions. Built in a weekend. This is the future of software.