If You Are a CoreWeave Shareholder...You NEED to See This Quarter

Welcome back to another episode here at What the Chip Happened, the greatest in my opinion semiconductor and AI channel on YouTube right now. Now, on today's episode, I want to take a closer look at Core Reef. This is a company that just reported earnings after hours. The stock is up nearly 15%. And for those that are not familiar, this is my biggest cost basis positions that I've ever made in an investment. This is a six faker plus investment in my overall portfolio and it's one that I am extremely extremely bullish on even though the market is completely completely wrong on this player. So in today's episode what I want to do is first take a closer look at some of the bullish sentiments that I believe came out of this earnings and there were a lot and a lot of bullish sentiments. Then I want to end with something from the backlog. The backlog did have something interesting and I feel bears could have a field day with the backlog story, but I want to showcase why the bears are going to be wrong yet again. So, let's take a closer look in today's episode. But before we begin, if you're serious about semiconductor and AI investing, I break down major earnings and conferences inside my community. Institutional quality research built for retail investors. Get 33% off at whatthechipappen.com. All right, like I mentioned, today we're looking at Core. just report our earnings after hours. The stock is up roughly 15.7% sitting at $104. Now, this is a stock that I've been very bullish on. And if you are part of, if you've been following the channel, if you've been part of the kind of what the chip happened community, make sure to check it out, whatchip.com, you've seen that I've been heavily buying this name in the low 80s, in the 70s, in the 60s because I thought it was extremely attractive. One of my latest buys was on July 29th when we did have those low 60s. Now for me is I believe the AI market a very top level right very top level if you're new for me I believe the AI market is a revolution revolutionary technology is going to completely shift how we do everything and for that reason I believe we are going to be completely compute constraint I don't think that any of the optimizations happening from open- source models from uh from tokens becoming uh cheaper from GPUs getting better and all these optimizations I think are great but it's still not enough to outstrip the amount of demand that we're going to see from AI due to various markets like cyber security like life science like AI agents like software solutions and coding and the list goes on and on. So very top level I am very very bullish in the AI space and I think anybody who has a lot of AI infrastructure like coreweave can benefit dramatically. Now I want to jump into kind of one of the big points uh that every bear in the AI story has tried to make. Every bear has tried to make that Nvidia GPUs as they release on an annual cadence basis is going to make older GPUs extremely extremely obsolete. But we continue to get the opposite answer. So this is what Cor CFO mentioned. He mentions that right now they're seeing upside of recontracting and that is largely going through their whole Nvidia GPU SKs from older generations to newer generations. And they mention as the older generation start to roll off their original contracts, they offer a strong return in delivery because for those that are not familiar, we're going to look at the next transcript in a bit, but when Coref makes a product and when they make these AI data centers, they pretty much expect them to be paid off after the first contract is over. And that's important because if you're getting recontracts after the original contract, that's big money coming in. For an example, they signed an A100, a contract that extends into 2029 at an attractive price. As a reminder, this Q was introduced in 2020. So, one of their GPUs, one of the older GPUs, and is extremely obsolete based on the Bears is now contracted to 2029 at attractive prices. So, this is a product that they expected, they it paid off its due, right? It paid off its due with the original contracts and now it's doing even more because the CFO mentions that they have built a business whose economics do not rely on recontracting after initial customer turn. Increasingly we are seeing longer utilization at higher prices offering the potential for significant further upside. So for coreweave to be successful and this is a massive bearish thesis right many people go core has this massive amount of debt they are just building and building but the way they build these contracts they are long-term contracts they build that capex and get that debt at the front end but by the time that contract is over after the five years that debt should be paid off the money free cash flow from that data center is money on their side and they end up with really great margins but Now, the upside is after that 5-year contract is done, the pricing of these GPUs are increasing, so all that money that they're going to be making is much higher than they thought before. Because remember, when you kind of used to make the modeling of AIG GPUs or any GPU or NI infrastructure, you would have this model where the GPU pricing would increase dramatically over the few years because of the apprec the GPU being weaker and not needed as much. So you had to put price it accordingly. But instead what you're seeing now is at this point where you thought GPU pricing were going to be down here by year five, it actually it's going to be up here and it's going to dramatically change the margin impact and that's one of the main reasons the company did see an increase in margins. Uh so margins did have a great great report this quarter as it was the bottom last quarter and from now we should continue to see growth and growth from here. The other thing that Cororeef is doing, right, and this is very exciting, like I mentioned, they really focused on long-term contracts at first to build this massive, massive fleet. And as we're going to look at some of the images, right, now they have 1.5 gawatt of active power. Out of this 1.5 gawatt, they added 500 gawatts of active power this quarter alone. This quarter alone, just to put it in perspective, they added more than every other NeoCloud has active right now. More than Nebus, right? Nebulas I think has like 300 and something active power that it's had for the past year or so right so it showcases the difference in size now the reason they were able to do build these massive infrastructure is because they have those long-term contracts but now they are changing the financing the financial side is getting better the terms for AI infrastructure is becoming better for the lender um we also have companies like Nvidia having some form of say that's saying look Jeep CPUs are going to be a strong revenue asset. So you tend to give these products a better term rate and we kind of saw it from and correct that the A100 an almost 7 six year old product is still generating revenue after. So all this is making the terms easier for a company like Corore and other players. Now they just mentioned that they had a new financing DDTL 5.5 and this opens up a new market for them. the clients that they are signing in with this type of contract right here to build this data center. They're only looking to buy compute for two to three years. Now, two to three years and shorter term contracts tend to have higher margins. So, Coree is now entering this new market for them where they're focusing not necessarily on spot pricing, but shorter contracts. We've seen kind of SpaceX do this. We see obviously Nebas do this. But with these shorter contracts, you tend to have higher margins. The risk is what happens after this three years if demand isn't crazy. But Corey believes demand will continue to be crazy three years after that that they will be able to re contract the shorterterm contracts with higher margins again in two to three years. And this is where the bulls and bears need to pick a side of where they believe this market is heading to. uh another kind of margin expansion outside of pricing outside of their solutions getting bigger um outside of just uh favorable financing as well. The fourth reason is they do have other solutions that they stack on top of just the bare metal um infrastructure play, right? They have a lot of things like storage solutions like uh security solution uh like software stacks on top of it and all of this is additive additive to their overall market they and margins. They mentioned in the past few months they launched kind of an inference-based product. it from 1 million it grew to more than h 100 million and now they expect this year to be at least $250 million in manage inference ARR now this definitely could be higher but remember a lot of the infrastructure that core has is for those long-term contracts and when their CEO and when management and CFO sees that look a company a contract is ending soon what are we going to do with this GPU do we take the safer route and still get great margins by relocating for a very long-term contract again or do we take our chances and do some of the spot pricing? So, they have to kind of balance it out. To me, even if they continue to do long-term contracts, this would be a very very healthy business for them. Obviously, they're trying to manage a little bit of growth and increase margins dramatically as well. And they are going to offset some stuff for lower uh for lower contracts which tend to have higher margins over time. Um we tal showed this slide already but we saw massive 500 megawatts of active power added. They increased their contracted power to 4.2 gawatt as of August 11th of 2026. They are expanding internationally. Personally I would love more uh I would love and continue to see the majority to be US-based. Um but it is nice to see that international expansion. Now let's talk about that backlog. Right? that backlog is is something that I think many bears might have um field day about. Uh but before we go there, if again if you haven't check out what the chipappen.com this is my personal community we do exclusive live streams Monday, Tuesdays, Wednesdays and Thursdays, three days a week. We have exclusive deep dives on earnings. We have a lot of different excuse me information around the whole industry. So, if you're curious about semiconductor and AI and want research at a great price, make sure to check out what the chiphappen.com. Now, with the revenue backlog, right, let's end this episode here. Bears are going to say two things. The first thing bears are going to say is, wait, revenue backlog was only $104 billion. Weren't you at hundred billion last quarter? You only grew $5 billion on a sequential basis. I didn't see that. That's definitely true. But they do mention that 25 billion is not included in this backlog that were committed in early quarter 3 of this year. Um or or or early quarter 3 which is where they're at right now. So few of the reasons that I believe backlog didn't see much growth sequentially is they don't have enough compute. they don't have enough infrastructure and they can't rent out uh it's impossible for them to say look you want to come in with me um and you're not going to get your GPUs for for the next 24 months obviously you're not going to make a contract for that uh we also saw that a good portion of the backlog even that 20 uh extra $25 billion most of it is going to the further end of the side it's 25 plus 25 months plus on when that revenue is going to be collected for the newer uh for the newer contracts and the newer backlog. Bears would say, "What's going on here? Why are you not getting that revenue earlier? It's a compute constraint market. You can only build so many GPUs at the beginning. And if you don't have the GPUs, you can't serve your customer." So for them, they already have a nice backlog that they're able to meet the revenues that they need for the next 24 months, which is 2 years. Obviously, you knew that, right? I don't know why I said that. like I made up like I just shared some revolutionary um status 24 months equals two years. Um but yeah, I think it's just compute constraint. So if they continue to accelerate their AI infra infrastructure investment, I think it's going to be a great time even though the market obviously hates capex, right? They did increase capex uh for the fiscal year. The market did see a dip after that, but um I I think it's needed. I am very much bullish in the AI market. Obviously, I understand there's bearish reasons, but when you when you kind of get into my head and understand that I am I believe the demand of AI tokens is going to increase dramatically and much higher than any optimization happening in the whole industry. I feel extremely comfortable owning a company like Corv. And I understand that the debt is not just built on hopes and dreams. It's built on five-year contracts and contracts that are expected to um be paid off in profits by year three or four. And now you're seeing added it renewal on a business that they don't need that. They would make money without that renewal contract. You feel comfortable. I feel comfortable owning a company like Cororeweave. When you're seeing pricing and of of A100s going higher and making contracts of a 60-y old GPU and adding three more years on lifetime, you feel I feel comfortable owning Core Reef. Uh and and when you're seeing core we've built so massive in active power where they're now 1.5 gawatt and it allows them to now no longer worry about long-term contracts and allows them to test the waters on shortterm contracts because they have this amount of massive amount of revenue coming in. I feel comfortable h holding corore reef. So I'm very much bullish in this company. I do believe this is a player that's going to be hundreds of billions of dollars of market cap and it's not going to be a straight line today. I probably am going to get a lot of love in the comments for Corore Reef, but who knows, a month from now, I might get a lot of hate like I was getting just a few weeks ago. So, take care. Have a good day.