# The Programmable Data Center — Evan Conrad, SF Compute

Markdown copy of https://realalpha.app/v/EAtLq9_vrE0/ on Real Alpha, the best minutes from long interviews with the people building AI. Cite and link the page, not this copy.

Evan Conrad says compute markets need independent operators and physical settlement to price real GPU capacity.

- Show: Latent Space
- Hosts: [Shawn Wang (swyx)](https://realalpha.app/people/swyx/)
- Guest: [Evan Conrad](https://realalpha.app/people/evan-conrad/) · SF Compute
- Published: Oct 20, 2026
- Length: 37 min

## The best minutes

- [9:35](https://realalpha.app/v/EAtLq9_vrE0/?t=575) [Compute contracts must price interruptions and hardware differences](https://realalpha.app/s/compute-contracts-must-price-interruptions-and-hardware/) (5 min)
- [14:13](https://realalpha.app/v/EAtLq9_vrE0/?t=853) [Batch and live inference need different interruption terms](https://realalpha.app/s/batch-and-live-inference-need-different-interruption-terms/) (5 min)
- [23:23](https://realalpha.app/v/EAtLq9_vrE0/?t=1403) [Financial compute markets need physical settlement first](https://realalpha.app/s/financial-compute-markets-need-physical-settlement-first/) (3 min)

## Also worth watching

- [28:58](https://realalpha.app/v/EAtLq9_vrE0/?t=1738) [SF Compute operates clusters without owning the GPUs](https://realalpha.app/s/sf-compute-operates-clusters-without-owning-the-gpus/) (3 min)
- [6:58](https://realalpha.app/v/EAtLq9_vrE0/?t=418) [A flexible order book can precede standard compute contracts](https://realalpha.app/s/a-flexible-order-book-can-precede-standard-compute-contracts/) (3 min)
- [20:17](https://realalpha.app/v/EAtLq9_vrE0/?t=1217) [Compute buyers can choose broad or exact GPU SKUs](https://realalpha.app/s/compute-buyers-can-choose-broad-or-exact-gpu-skus/) (3 min)
- [26:51](https://realalpha.app/v/EAtLq9_vrE0/?t=1611) [Compute trading needs a supporting financial ecosystem](https://realalpha.app/s/compute-trading-needs-a-supporting-financial-ecosystem/) (2 min)

Skip the event and podcast catch-up between the opening market question and SF Compute's status update.

## Summary

Conrad says SF Compute began as Junark, a company that tried to sublease its cluster. He says the company then built the cloud operating capacity needed to deliver compute reliably and now offers the supercomputing side of GPU clusters without owning the hardware.

Conrad argues that GPU contracts need to reflect the real constraints of delivering compute, including workload interruptions, storage, network links and hardware differences. He says SF Compute built cloud operations and a flexible order book to support physical settlement, while leaving capacity owners independent of the market. He favors establishing physical delivery before building cash-settled contracts, which he says could otherwise track prices disconnected from available capacity. He also describes why large GPU suppliers and cloud providers may lack the independence needed to run such a market.

### Minute by minute

- [2:08](https://realalpha.app/v/EAtLq9_vrE0/?t=128) **SF Compute grew into cloud operations.** Conrad says SF Compute began as Junark, subleasing its cluster. Delivering compute reliably meant taking on the operating work of a cloud, which he compares to selling a hotel room and then adding housekeeping, room service and security.
- [5:31](https://realalpha.app/v/EAtLq9_vrE0/?t=331) **Liquidity is concentrated on limited capacity.** Conrad says the order book is still small, with 5% of SF Compute's capacity on it, but has concentrated liquidity. He tracks whether customers can resell capacity: a customer spending $100 to buy capacity makes back $25 on average by selling it to someone else.
- [7:10](https://realalpha.app/v/EAtLq9_vrE0/?t=430) **The order book handles nonstandard compute.** Conrad says standardized contracts remain a long way off because compute operations and performance vary. SF Compute matches orders across properties such as hardware, time and performance; this supports trading before standards exist, but leaves wider spreads and less concentrated liquidity.
- [9:53](https://realalpha.app/v/EAtLq9_vrE0/?t=593) **Workload interruptions depend on infrastructure.** Conrad says buyers need to know whether they can tolerate interruption and what storage and network links support recovery. Fast storage and transit between clusters can let workloads move and resume elsewhere, but many GPU clusters lack those connections.
- [11:53](https://realalpha.app/v/EAtLq9_vrE0/?t=713) **Hardware cohorts affect reliability.** Conrad says customers may pay more for later-deployed GPUs because early H100 and GB200 systems had problems. He distinguishes those issues from burn-in testing, citing firmware bugs in a system's components that could require updates or replacement.
- [14:22](https://realalpha.app/v/EAtLq9_vrE0/?t=862) **Contracts fit different workloads.** Conrad says clusters are financed over three to five years, while order-book capacity can range from the next hour to five years out. Batch inference can tolerate short-notice interruption; live inference may need one to three months to adjust, so the two types of interruption carry different prices.
- [17:55](https://realalpha.app/v/EAtLq9_vrE0/?t=1075) **Compute prices form a hump.** Conrad says near-term capacity that has not sold can drop in price to clear, while future capacity may be worth less because buyers prefer compute sooner. He says an inference-running B300 could earn about $40 an hour at full utilization, or $25 at 50% utilization, while B300 capacity can be bought for $5 to $6 an hour.
- [20:17](https://realalpha.app/v/EAtLq9_vrE0/?t=1217) **GPU generations remain distinct.** Conrad says customers care about chip generations because specifications such as VRAM and NVLink domains matter. SF Compute lets buyers define what they will accept, such as Blackwell capacity at $6 an hour or less, while more specific customers can request and pay for exact hardware.
- [23:23](https://realalpha.app/v/EAtLq9_vrE0/?t=1403) **Physical settlement limits basis risk.** Conrad says the industry should make compute more fungible, but warns that financial contracts created before physical delivery could be easy to manipulate and expose buyers to basis risk. He says cash settlement could be useful later if it is derived from physical settlement and real transaction data.
- [26:51](https://realalpha.app/v/EAtLq9_vrE0/?t=1611) **Compute markets need an ecosystem.** Conrad names Silicon Data and Compute Exchange as early participants pursuing cash settlement, which he says can coexist with SF Compute's physical-settlement approach. He describes a broader ecosystem forming around standards, residual-value insurance and deposit lending to address compute's financial bottlenecks.
- [28:58](https://realalpha.app/v/EAtLq9_vrE0/?t=1738) **Physical settlement requires cloud operations.** Conrad says SF Compute visits data centers, helps design bills of materials and clusters, coordinates supply chains and manages the equipment around the GPU nodes. The company partners with data-center owners rather than owning the hardware, which Conrad says helps it remain independent.
- [32:18](https://realalpha.app/v/EAtLq9_vrE0/?t=1938) **SF Compute supports sovereign clusters.** Conrad says the company deploys clusters around the world and works with local operators that want to become clouds. SF Compute brings its expertise, helps set up the clusters and aims to make them financeable.
- [33:19](https://realalpha.app/v/EAtLq9_vrE0/?t=1999) **Conrad says incumbents cannot run an independent market.** Conrad argues that NVIDIA, hyperscalers and GPU cloud providers cannot operate an independent market while pursuing major inference and cloud deals. He says independence can support liquidity and lender financing against the forwards market, lowering financing costs across the industry.
- [34:41](https://realalpha.app/v/EAtLq9_vrE0/?t=2081) **The data center becomes programmable.** Conrad says SF Compute has closed several hundred million dollars of volume and has more clusters on the way. It lets customers subdivide their InfiniBand fabric on the fly, and he describes the goal as virtualizing the data center; cluster owners can retain the hardware while SF Compute runs the supercomputing side and owns the SLA.

## Transcript

Latent Space, hosted by Shawn Wang (swyx), with Evan Conrad.

Transcribed by machine (Whisper) from the episode’s audio, so a word may be off here and there.

[0:00](https://realalpha.app/v/EAtLq9_vrE0/?t=0) NVIDIA is obviously like the 800-pound gorilla in the room. Why didn't they make SF Compute? So our impression is that NVIDIA can't make it, but benefits tremendously from it existing. The reason why they can't make it is because they would then not be independent enough for people to trade on it. Anyway, Evan, welcome back to Lanespace. Thank you so much. Welcome back to AIE. Actually, first time at AIE. I think so, yeah. This is a very special one because I feel like you have been working on compute trading, compute markets for a while. Now it's like feeling of the moment. Yeah. Jump basically announced that they are getting into compute markets, which is why I'm excited to sort of cluster people.

[0:49](https://realalpha.app/v/EAtLq9_vrE0/?t=49) You met Oliver from Semi Analysis, who is basically initiating coverage on compute markets is my read on this. um so people are going to watch your talk i think it's very good explanation of like both your origin story and why you need to exist and what the hard problems you work on yeah uh done in a very simple engaging manner i'm very jealous of your writing skills because you showed the flyer for that you showed at new reps yeah i was like it's pretty straightforward and like simple and like yeah it's it's good yeah so anyway congrats thank you i appreciate that um and also congrats on AIE, it seems to...

[1:24](https://realalpha.app/v/EAtLq9_vrE0/?t=84) Y'all are rolling. We're rolling. I just came back from the site tour for next year's AIE. And then for... Like a hint? At Javits Center. It's the same place they do Comic-Con. Okay, wow. Wow, wow, damn. But I think the finance focus is really hitting for folks. Yeah. So we're going to try to split out. Like this is... Here, we just lump everything in finance. Like your finance, their finance, like insurance is finance. But like they're all super different. Yeah. So we'll probably break out things more minutely in future because people want to go deep on hedge fund stuff or banking or personal wealth.

[2:00](https://realalpha.app/v/EAtLq9_vrE0/?t=120) Yeah, especially in New York. Yeah, like Anthropic was like, yeah, we're all about RIAs right now. RIAs, like Registered Investment Advisors. Yes. Like so different from you. No, no, no, I'm aware of this. Yeah, yeah. Okay, so for listeners who don't know, we did a podcast two years ago where you talked about where SF Compute was. Today, you're much more in this sort of physical settlement. You talk about your back fabric. What do you call it? The back plane. Back plane, which obviously takes a lot of hardcore engineering stuff. Yeah. Can you give a status update on where is SF Compute versus listeners who last heard you on a podcast two years ago?

[2:40](https://realalpha.app/v/EAtLq9_vrE0/?t=160) So when we started, we were this company called Junark, and then we were just trying to sublease our cluster. um and we started there in the technical stack um and a way to think about it is it's like you're in a hotel and you're trying to sell that hotel room and then somebody asks you hey like are you gonna clean my room and then you say oh i guess i probably also need the hotel room as well or like i need the um the cleaning closet and so you go get access to the cleaning closet and then they want room service and so you get access to the kitchen and then you know they want um security and so you get access to the security closet and you get this and you do that and you do this and at a certain point you're just running the whole hotel.

[3:17](https://realalpha.app/v/EAtLq9_vrE0/?t=197) And the moment you're running the whole hotel you say, oh, wait a second. Instead, my supplier on this market is somebody that looks closer to a fund or a cloud that has access, in our case, a cloud that has access to the data center and we're going to level up. And so what we've spent the last several years doing is building the entire operating capacity of a cloud in order to actually deliver the compute at a reliable level. We call that physical settlement. We're taking a term from the futures market, from futures markets, from other markets, where in oil or soybeans or something, physical settlement means you get delivered the oil and the soybeans yes um but um in some markets um like in oil and also in power um there's like a delivery network in oil you have to have pipelines if you would like the oil to be delivered to you there has to be a way for it to get to you yes um and finance and for people interested in financial history during covid oil went negative yeah because uh so like i will pay you to take oil from me because it is sitting there in the the port uh and like i can't pay for this yes and so i will pay you to take my oil yes that is like the weirdness of physical

[4:37](https://realalpha.app/v/EAtLq9_vrE0/?t=277) settlement yes there's a bunch of weirdnesses like this in other markets as well so in the power markets and the grid um there's parts of the orders that you fill out the buy orders and accelerators that are there because the boiler takes a while to cool off. And you can stop the boiler and then you can start it back up again within some period of time. And so that has to get factored into the order book. This genre thing is the same thing that happens in compute markets. So we co-designed physical settlement and the order book at the same time. And that meant that we get to actually handle the real constraints of compute, which might include something like, oh, I'm okay with being interrupted, but only if there's storage on the thing that I can recover from really quickly. Very similar types of problems. Yeah. So people can refer to your talk, and you've done a couple other podcasts already. So I want to be the sort of 201 level instead of the 101 level. And also because I have been a derivatives trader, so I can give people a bit more detail that you normally don't get into. So for example, how mature or deep or liquid is the compute market today?

[5:46](https://realalpha.app/v/EAtLq9_vrE0/?t=346) Not very mature or deep or liquid. So what we've been doing is first we had to go get a bunch of supply online in the first place, which just takes a while. And then we had to build the entire platform in order to do this. And we just now got that working. And so now we move capacity over onto the order book, which makes it liquid. And so 5% of our total capacity is on the order book. And we're consistently moving more and more. And then we're bringing on more capacity that's all on the order book. The nice part is that maybe I should have said the volume on the actual order book is small, but the liquidity on the order book is high.

[6:23](https://realalpha.app/v/EAtLq9_vrE0/?t=383) So for that small amount of capacity that you have, you have like very concentrated liquidity. So a sort of standard metric, I would say, for someone who is using this is a customer who buys capacity from us and they spend $100 to buy that capacity will make back $25 on average from selling it back to somebody else. And that's the number that we, or this is one of the metrics that we try to track, because we want the liquidity to be useful to the commercial participant. And you can measure liquidity in a bunch of different ways. You can say, like, oh, I just want tons of makers or something. And you can measure it in, like, high-minded ways.

[6:58](https://realalpha.app/v/EAtLq9_vrE0/?t=418) We try to measure it by, like, is this solving the problem for the customer? Yeah, got it. And so last time we talked as well, there was a lot of conversation about standardizing the contracts. Yes. Talking with like the CFTC or whatever, what's the status of those things? So I think standardizing the contracts in general is still pretty far away. And what we were trying to do at the time was, or basically what we learned over the last several years was we just needed to operate the whole thing. Otherwise you cannot standardize the thing. And so to get closer to something that can be standardized, you first need to make all the operations of the thing the same.

[7:39](https://realalpha.app/v/EAtLq9_vrE0/?t=459) To give an example of what can be different is a single operator can offer two different types of virtual machines. One type of virtual machine might have a 40% performance degradation versus another type of virtual machine. And 40% is a lot. When you say performance degradation, are you talking like MFU type things or like amount of... burn-in that you've experienced, or what? It will affect, so in this particular case, I'm talking about GPU kernel launches and how long it takes for that to happen, which is a process that happens quite frequently when you're doing sort of either training or inference.

[8:14](https://realalpha.app/v/EAtLq9_vrE0/?t=494) The, that will basically, the end result is that you get a worse MFU, but like, it's the intermediate level. Yeah, it's the intermediate level, basically. Yeah, so what I'm trying to, so for anyone who's traded commodities before, there's different grades of commodity, right? There's like, it's not oil, it's light sweet crude versus like whatever the opposite of light sweet crude is. Or it's like, you know, corn level A or corn B or whatever. And so people are trying to, well, you have to establish the equivalents of those things. Correct, yes. And I'm wondering if they have been established. So what we did instead was we created an order book that had multiple different constraints.

[8:47](https://realalpha.app/v/EAtLq9_vrE0/?t=527) And you can say, I want to have, you know, something that's between this and that. It runs from this time to this time. it's like x to y it's a bunch of different properties that you can attach your order and then we match that against cell orders that have those properties in a big matching engine

[9:02](https://realalpha.app/v/EAtLq9_vrE0/?t=542) **Evan Conrad:** and the reason why we did it this way is it allows us to generate liquidity without having to standardize before you know what the technology can do and because the technology was shifting so

[9:12](https://realalpha.app/v/EAtLq9_vrE0/?t=552) quickly we wanted to basically be able to do that the trade-off of this is that you have less concentrated liquidity and so the prices are going to have there's going to be more spreads between the the buy and the sell yeah it sucks yeah but on the other hand it means there's lots of arbitrage to make um if you can do that correctly so if you have a product that can use lots of different types of capacity um then you might really like sf compute because you will be able to use lots of different types of capacity and you'll make more money um you can take the spread okay what dimensions matter in this grid of things um so a bunch of different things um one is dimensions categories that i will say is like is there a workload running on it right now or is it capacity in the future so is it capacity right now um capacity right now in order for you to sell that um you have to interrupt your current workload so you have to think do i care if my current workload is interrupted or not capacity that's sold that's starting a month from now um you can buy and sell that all the time and so you get um liquidity on that really easily um there is some concern like oh no well i have capacity next month but it's very different than if you're interrupting

[10:18](https://realalpha.app/v/EAtLq9_vrE0/?t=618) a workload right now the way that the interruption happens matters so an example is if you're on a cluster and the cluster doesn't have as amazing storage as you want it to have so an example is like do you have a weko or vast system under the hood and do you offer that and like what's the size of that this changes how you respond to being preempted or removed off the cluster meanwhile the location of that storage matters so if you're able to move that data from one cluster to another um quickly then maybe you might be able to spread across multiple different clusters um and you can be interrupted over here and then pop back up over there um to make that work you also need um like a fast transit path between one cluster another typically this is implemented with a wan but most of these clusters don't necessarily have um like a wide area network yeah a wide area network so um basically cables that connect the clusters together a private internet is another way to think about it um a lot of clusters that are being spun up today They're not attached to traditional CPU clusters were next to the customer for latency reasons.

[11:27](https://realalpha.app/v/EAtLq9_vrE0/?t=687) In GPUs, they're next to the power, wherever you put the data center next to the power rather than next to the customer. This means that a lot of stuff is getting put in places that previously they weren't expecting data centers there. And so you may not have as excellent of a transit path from one to the other so this this starts to matter um other examples are um weird shit where we theoretically support but don't actually um uh like this is a thing that we're thinking about haven't done basically um the early gb200s broke the early h100s broke um and so if you were buying h100s or you're buying gb200s you might really want the ones that were deployed after a certain time because you're not getting the first shitty ones off the rack and so if you're a super sophisticated customer you may be willing to pay more um because for that oh so it's it's about like what cohort they were they were produced yeah and they get better over time so like uh the first ones off the rack maybe had some faults or some problem. Right. And so if you were the ones who got the first ones, or, you know, first ones from the first shipment of them, then yeah, they would be broken or in some way. I assume the sort of

[12:42](https://realalpha.app/v/EAtLq9_vrE0/?t=762) process is, like, look, there's defects everywhere. Like, it just happens, right? Totally. And so, like, you get a new shipment from TSMC or whatever, and you just kind of run your tests and, like, you figure out which ones are bad and you throw them out. And isn't that how it works? I didn't know about this element of cohorting where like let's say Q4 2026 B300s are way worse than 2027 B300s. Yeah, so I'll give an example. You're describing burn-in. Yes. So when you get a cluster,

[13:14](https://realalpha.app/v/EAtLq9_vrE0/?t=794) **Evan Conrad:** a good cloud, what it does is it tries to kill all the GPUs before they ever get in the hands of the customer. And then

[13:21](https://realalpha.app/v/EAtLq9_vrE0/?t=801) the GPUs that are broken or the nodes that are broken, you get RMA'd. You basically go and request it. you return it and you get a new one and there's a lead time to that but whatever you'll get yeah you'll get some back um and then the customer never has to experience this um the um thing i'm referring to is that um like in early h100s there were like firmware bugs that may not even be related to the specific like they may not be NVIDIA they might be like super micro has a firmware bug in the um in the box that you bought um and it applies to all of them and it doesn't matter like um the solution to that was you need to get them to update their firmware um or they need to ship a new thing um in every new box yeah well commission had this for like the the b300s that that we got um yeah and the iq had to use devin to go fix it yes so like different types of problems like this can happen interesting okay so i'm like okay that seems small compared to the overall like i just want to like i want people to understand the shape of this market um you mentioned like preemption you know and and you know we talked about spot instances and how that the analogy is yeah and then there's also you know booking a month out how far out do this do

[14:34](https://realalpha.app/v/EAtLq9_vrE0/?t=874) these markets go and is is real time or sort of preemption more interesting to you versus uh far being further further out um to answer your first question um three to five years um wow we're And like month by month or quarter by quarter or how's this? Are you asking how the cash flow works? The contracts, like the futures contracts, right? Again, I'm a former contracts derivatives trader. In the market itself, like is in the broader market, clusters are financed three to five years. So the payments are for over that period of time. There is also like blocks of capacity in our order book, which can extend any arbitrary time.

[15:14](https://realalpha.app/v/EAtLq9_vrE0/?t=914) There's some stuff that we have to do for the cash flow on that. Right now what we do is we just ask people to pay entirely up front for a block. We don't ask them to do that for a lease for a longer term contract. This gets into something that I definitely need a whiteboard to explain. But you had another question I was about to answer. Do you want real time? What's more interesting to you, real time or longer out? So we care more about longer out because we're trying to build a forwards curve that has a lot of liquidity on it. the, these are kind of the same thing and kind of not. So what we do is we can say, you can buy an hour out.

[15:56](https://realalpha.app/v/EAtLq9_vrE0/?t=956) You can buy right now to the next hour, or you can buy right now to. the next three years and anywhere in between that's wild i know um and so the um you know or five years technically um the there's a certain difference that happens that's dependent for each customer basically like if i have a workload that is um uh that's batch inference basically where i can be interrupted at any point i don't really care um then my ability to be cut off is probably like, I need an hour of time to know that I'm going to be cut off and then I'm

[16:37](https://realalpha.app/v/EAtLq9_vrE0/?t=997) **Evan Conrad:** good. Or I can even be notified in, you know, a minute or two. But if I have live inference customers who are expecting me to be online, then you can interrupt me, but in a month, maybe two months, maybe three months. And then I can adapt to that capacity over that period of

[16:54](https://realalpha.app/v/EAtLq9_vrE0/?t=1014) time. But you cannot interrupt me in a minute. And so those two things are priced differently. And the reason why they're priced differently is because what is the price you would pay in order to interrupt your batch job? It's very different than the price you would pay to interrupt your live job. If you have a live job, you are probably paying an SLA when you break for your customer. Does that make sense? Yeah, I see. So this is actually the market for breaking leases. These are the market for breaking leases, basically. Oh, my God. And then there's also the market for future leases, or like for leases that are...

[17:29](https://realalpha.app/v/EAtLq9_vrE0/?t=1049) Oh, I thought this was like a subletting situation. This is like... Oh, the lease break. So one way to think about it is in power markets, there's actually two markets. There's like the day ahead market and there's the live market. Compute kind of has the same thing, where it has capacity that is coming online, or capacity that's about to arrive to you, and then capacity that is currently in use. And these have different price dynamics. Okay. Interesting. Backwardation or contango? Backwardation or contango? So is compute further out price higher than spot or lower than spot? Is compute further out price?

[18:10](https://realalpha.app/v/EAtLq9_vrE0/?t=1090) Does it go like this? Does it go like that over, you know, over like one year out, two year out? Do? Oh, this is both at different times. I find it fascinating that you don't know. Yeah. I mean, both. It depends on who you are. Okay. So I'll make the case – wait, so you're asking, is compute sold right now? So compute like the same unit of compute today, maybe not today, maybe a month out versus one year out versus two year out. Is that curve going up or going down? We see like a hump. Oh, my God. Okay. Explain. Like I think this will make intuitive sense. It's basically like if I have milk and it is about to expire, I will lower the price on it.

[18:48](https://realalpha.app/v/EAtLq9_vrE0/?t=1128) Yes. And so that's basically what we see. So it's the NVIDIA release cycle, basically, right? Well, we're more saying, like, if there's a block of compute and it's approaching now, it can't be sold in the past. And so it has to drop its price in order to clear at some price. And so initially you see this, like, drop where stuff that's sold right now, anything that's on the market right now, is something that hasn't been sold. And so therefore the price will drop. And then it goes up to where it wants to be. And then maybe compute in the future is worth less because people don't want – like people would like the compute now.

[19:22](https://realalpha.app/v/EAtLq9_vrE0/?t=1162) And so they're sort of paying for the shorter lead time. And they pay less for the compute that's off into the future. This is what I think a normal healthy curve looks like. The place where I'm like, oh, man, no shit, is that right now if you took a B300 and you tried to run inference on it, you can make $40 an hour or so selling that. any model basically um and so that's if you use it at 100 and maybe if you only have like 50 utilization or something so like 25 an hour yeah so 25 an hour meanwhile you can buy v300s at like five to six dollars an hour one of these numbers has to change. Yeah, it's a huge overcharge.

[20:02](https://realalpha.app/v/EAtLq9_vrE0/?t=1202) It's a huge overcharge. You mentioned this in your talks a lot. Yeah, so a bunch of people are basically looking at this and thinking, if I think prices are going to go up, I might be willing to pay more than what I previously thought in the future, because I might think, OK, I can buy a five right now. But if I can't buy any more, then maybe I will be willing to buy even more in the future just to lock it in if it's going to come online later. So we see the other side as well. Kind of depends on who you are. Cool. So, is it, okay, and I just really want to make super clear on this. Of course.

[20:33](https://realalpha.app/v/EAtLq9_vrE0/?t=1233) Do you care about the different generations of chips, or do you just care about overall flops and you spread it among whatever chips you want? We care about different generations of chips because... Yeah, you care. I guess, do your customers care? Yes. Right? Oh, wait, when you say different generation chips, do you mean like... So, like B300s or B300s? Yeah, exactly. Yeah, or like even AMD or whatever, right? Yes, I think people totally care about the generation of chips. I think part of this is because of things like VRAM, and part of this is because of larger NVLink domains and things like that.

[21:04](https://realalpha.app/v/EAtLq9_vrE0/?t=1264) So there's just no way to abstract all this to some generalized unit of compute that you can sort of mush around, whatever you like. I think when you were saying even oil isn't oil, oil is actually just a bunch of different types of oil. Yeah, like jet fuel or car fuel, whatever. I think it's the same thing. You're going to get a bunch of different types. It is, but you have oil refineries. Effectively, a black box is you can put oil in then get like whatever out um if compute it is so uh heterogeneous i guess is the word is that you just have to expose the underlying complexity to your end customer which is bad for you as a market maker maybe like you want to identify the one like choke point or like i don't know what the better word for choke point is like a shelling point where everyone agrees like okay like this is kind of the standard and we'll we'll all agree to this standard and uh you know it'll be somewhat inefficient yes uh so we've been able to reduce the um the the problems to a certain level but yeah like your job is like dimensionality reduction yes we reduce a lot of the dimensions but nonetheless there are still dimensions yeah it's still going to exist do you know how many

[22:10](https://realalpha.app/v/EAtLq9_vrE0/?t=1330) skews you have so um like five times a hundred times like five thousand equals i don't know like what's we actually made skews like a concept um that and the goal is to make it so that

[22:26](https://realalpha.app/v/EAtLq9_vrE0/?t=1346) **Evan Conrad:** somebody else can make their own skew so if i want to buy blackwells i can say i'm willing to buy blackwells at six dollars an hour or less and you can match me against gb300s or you can

[22:38](https://realalpha.app/v/EAtLq9_vrE0/?t=1358) match me against b300s and i don't care which blackwell as long as it's less than six dollars an hour um but you just said you care about memory and other stuff totally but not every customer does so some customers like um just can work on black holes whatever it is some customers are super hyper specific and they need this exact thing and they're going to pay more for that because the market's going to charge them more okay um and that's what we do so and so people use your apis to programmatically define what they want and then you match that with whatever composition of sellers uh they might they might they might be able to fulfill yes um with the caveat that having a named query is a new thing we're doing um but the query system as a concept is how the order book works right now.

[23:17](https://realalpha.app/v/EAtLq9_vrE0/?t=1397) Okay, interesting. All right, so I think that's a lot of, I guess, nitty-gritty. Are there anything about, I guess, that you want to educate people on, whether it's financial people or engineers, about the state of the compute trading market as you've developed it? Because effectively you guys are defining it. Mainly just the requirement for physical settlement. We think that if you... We spent three years learning why you need physical settlement. And our entire conversation today is that these things are not as fungible as you think. We're going to try and make them fungible. That's the industry's job to do that.

[23:57](https://realalpha.app/v/EAtLq9_vrE0/?t=1437) But without that, you run a lot of... risks. If you try to build too much financialization before the industry has had time to build out the physical settlement, then I think what you have is you have a market that's very easy to manipulate and a market that has a lot of basis risk, where the price you may be hedging against actually has nothing to do with the price that the market you can actually buy the capacity for. Yeah. So for listeners, we talked about this briefly last time, but people are often shocked to find out that, let's say, the market for credit default swaps is 12 times the size of the bonds that they insure, something like that.

[24:40](https://realalpha.app/v/EAtLq9_vrE0/?t=1480) Or like the market for mortgage sort of reinsurance or all that is larger than the underlying physical settlement. So that's why physical versus cash settled. And basically cash settlements are exactly the kind of side bet that you see in the big short where they may or may not have anything to do with the underlying thing itself. And it may be used as a proxy for overall other stuff. And you don't want that to go out of control? Yeah. So to be clear, we think cash settlement as a future concept is a great idea. We just think it should be derived from physical settlement. because, as you said, it's a side bet.

[25:20](https://realalpha.app/v/EAtLq9_vrE0/?t=1520) And it would be kind of like if you made a cash-level future before you made the physical settlement, which I think people are kind of doing already. Yeah, totally. People are trying to make cash-level before you make the physical settlement. And people do that for all sorts of things, including startup stock. Yes. Right? Like, you can buy and sell a made-up share of OpenAI, but it's nothing to do with the actual share. Yes. And lots of those go very poorly. It would be kind of bad if you started selling the derivative on a stock that doesn't exist yet. Yeah. So, I mean, one thing that we do, so I used to be a currency trader, right, and interest rates as well.

[26:02](https://realalpha.app/v/EAtLq9_vrE0/?t=1562) And one thing we do is you have a board that gets together and says, like, this is the London Interbank Offered Rate, LIBOR. And we all agree, like, this is the monthly fixing, and we all trade to that fixing. Yes. Is that even meaningful at some point in three years from now? So critically, LIBOR was manipulated. Yes. Our concern at the moment is... Is there 12 banks? Yes.

[26:24](https://realalpha.app/v/EAtLq9_vrE0/?t=1584) **Evan Conrad:** Well, our concern is there's a very similar problem where you can manipulate an index.

[26:30](https://realalpha.app/v/EAtLq9_vrE0/?t=1590) Our impression is why you can manipulate that index is because a lot of the indexes are based on a very small amount of data that has not a lot of correlation with the actual price of buying a new reserve contract. That's what we're concerned about. Okay. Interesting. All right. I think that's the rough state of things. I think the other part of just today is you talked with Oliver, and you're doing an event with Jump, and you're friends with Silicon Data. What are you learning from all these other players? Who are the major players that you would want to call out as doing interesting things? So we're big fans of Silicon Data and Compute Exchange.

[27:13](https://realalpha.app/v/EAtLq9_vrE0/?t=1633) I think they were also very early to this, and I think they are doing a reasonable shot at the A path that we think that we'll likely end up being partners with. Can you distinguish your different paths? So the general list is they're doing cash settlement, we're doing physical settlement. We think both can exist and should exist, and that they're sort of mutually beneficial to each other, and so totally fine. So I think they are one of the earliest in the space, and I think they very clearly took the bet um well before anybody else did and i think that's like very honorable to do um um you mentioned uh semi-analysis jump like these are the players emerging like as far like you were my first introduction to this yeah and then and i see all these other guys coming in and i'm like okay like it's something's happening here i don't know the escape as clearly as you do yeah so um the categories of folks that we've seen is there's folks that are trying to set better standards so semi-analysis i think is working to try and create better standards um there's folks that are trying to do residual value insurance and so on um there's a number of different companies in the space i don't know any of them super well but

[28:18](https://realalpha.app/v/EAtLq9_vrE0/?t=1698) i know this is a concept and i think that's important um there are folks that are trying to do a deposit lending where like they'll front your credit risk for the cloud.

[28:29](https://realalpha.app/v/EAtLq9_vrE0/?t=1709) **Evan Conrad:** There's an ecosystem of companies that are reforming around how do we resolve the financial bottlenecks of compute?

[28:34](https://realalpha.app/v/EAtLq9_vrE0/?t=1714) And it is cool to see. I think effectively what you want in this ecosystem is buyers and sellers. You want an analyst. You want, we said insurance or what's the? Residual value insurance. Residual, yeah. And we want this healthy ecosystem out there. But obviously, you guys are owning the problem of physical settlement, which is really interesting. Okay, so we can go into the hard tech side, because obviously you are also hiring for Extremely Correct Engineers. Describe your stack. We do all the things a cloud does without being the cloud, as in we don't own the hardware. The reason why we don't own the hardware is because if we did, we would be able to manipulate the forwards market.

[29:18](https://realalpha.app/v/EAtLq9_vrE0/?t=1758) So we have to be Switzerland. We have to be independent, which creates a lot of sort of strange things about the company. But in order to do physical settlement, you have to do site visits. You have to visit the data center. You have to design the bill of materials and ensure that you have to basically do like supply chain consulting where you have relationships with different OEMs that ensure that things get delivered on time. then you have to do cluster architecture so you're sort of designing the cluster itself then you're rolling out racks or you're working with partners to roll out racks and supervising that then you're doing backplane management so you're taking over the PDUs and the switches and the DPUs and UFM and BMC and you're operating the whole harness around the nodes that you're selling when you say you as an SF compute employees yep wow So when you buy a node from a cloud, there's the bare metal.

[30:21](https://realalpha.app/v/EAtLq9_vrE0/?t=1821) But there's all this shit around the bare metal.

[30:24](https://realalpha.app/v/EAtLq9_vrE0/?t=1824) **Evan Conrad:** And in order to make that bare metal node work, you have to operate everything else.

[30:31](https://realalpha.app/v/EAtLq9_vrE0/?t=1831) Then we... Because I just like, traditionally the job of the data center is to extract some of this away from you. Yeah. We're the ones that do that. But you're not in the data center business. You're not in the sort of equinexes of the world. We're not in the data center business, sort of. What we do work with is – so we offer supercomputing as a service. And what this means is folks come to us and say, I've got X, where X is – I've got X megawatts here. And maybe sometimes they mean I've got a data center that's – like I would like to turn into a GPU cluster. Like my options are I could sell the data center to somebody or I could become the cloud.

[31:15](https://realalpha.app/v/EAtLq9_vrE0/?t=1875) And then we go to them and say, great, we will turn you into the cloud. We're the supercomputing part. You're the data center part. Let's partner together. And then we will sell and I'll make money. Sometimes it's people who have something that could turn into a data center, in which case we're going to basically project manage that up to the layer. and then we will work with those folks to turn their powered land or whatever they have into a data center by working with partners and then we take over at the supercomputing part. This requires us to have some sort of data center expertise in-house but we don't self-perform builds.

[31:53](https://realalpha.app/v/EAtLq9_vrE0/?t=1913) So, yeah. Yeah, so it's somewhat modular because you have different scenarios you need to play under. Yeah, someone once described us as like Palantir for compute, because it's like each time it's this new rodeo, and we deeply partner with folks on both sides. And primarily the US? Primarily the US, but we operate around the world. Yeah. At some point, does SF Compute become a misnomer? So we're deploying clusters all around the world. And when we do that, we're working with people who themselves are trying to be clouds all around the world. They look like sovereign compute players. And our pitch to sovereign compute is amazing.

[32:38](https://realalpha.app/v/EAtLq9_vrE0/?t=1958) Let's help you set up a cloud in your own country. And we will bring the San Francisco offtake. We'll bring the San Francisco expertise. We'll set up the clusters. That's our whole business. We're going to make it financeable. Who are your biggest announced partners that you can name? Nobody I can name, basically. we will have some announcements in a week or two that we expect to come on all things sure, sure, good luck I think this is really interesting for every country who's investing in sovereign AI to figure this out it's interesting that SF is still that brand name that is very marketable I think one of the interesting things is just thinking through this market um NVIDIA is obviously like the 800 pound gorilla in the room yeah um why don't they why didn't they make sf compute and again like i you know the answer is going to be market manipulation but like let's just really get into the incentives of the multiple players uh yeah so our impression is that NVIDIA can't make it but benefits tremendously from it existing the reason why they can't make it is because they would then not be independent enough for people to trade on it if you don't trade on it then you don't have enough liquidity for lenders to finance against

[33:54](https://realalpha.app/v/EAtLq9_vrE0/?t=2034) the forwards market yeah this is also why crusoe and core weave and lambda why the hyperscalers can't make it why base 10 and fireworks and modal can't make it because they have too much demand on that side um this is a thing that benefits the entire industry because it makes the financing costs go down but nobody can make it unless they give up the ability to do all the big inference money and all the big cloud deals um you have to kill your business in order to make this That's why nobody does it. It's like structurally, they're not set up to do that. Yeah. And you're basically reducing friction, and that's all your revenue.

[34:30](https://realalpha.app/v/EAtLq9_vrE0/?t=2070) You've announced some milestones. Actually, I forgot to say that up front. Can you repeat what they are and just take a victory lap a little bit? We've closed several hundred million dollars of volume. We have a significant number of new clusters on the way. Yeah. And then the things that I think I'm most excited about is we've shipped a whole bunch of new features that make SF Compute very viable for platform-as-a-service companies. An example is we allow customers to subdivide their InfiniBand fabric without having to ask us. So many clouds, when you go to buy them, they will give you one InfiniBand fabric.

[35:14](https://realalpha.app/v/EAtLq9_vrE0/?t=2114) And then if you want to redraw the lines of that Infiniband fabric, you need to contact them and they will then do something for you. We now can do that on the fly for customers, which I think is really great because it interfaces with the market really well, where you want to be able to redraw things in order to scale up and down your clusters. Would you describe this as like you're virtualizing the data center and making it programmable? Yeah, we're virtualizing the data center and making it programmable. So all the stuff that you would have gotten if you had owned the hardware, We're trying to virtualize that, which I think is exciting to us.

[35:49](https://realalpha.app/v/EAtLq9_vrE0/?t=2149) It's funny. The last time this happened was the cloud, which is really cool. And then finally, I think, what is your call to action for people? Is it engineers? Is it give me your supply? What is it? If you would like to own a GPU cluster and you want the residual value from it, you want to have it for your portfolio, we would love to work with you in order to build that GPU cluster. we will take over and we will do the supercomputing portions of that and ensure the things go right. We own the SLA. And then if it turns out that that was a bad idea that you need out of that cluster, we can immediately sell that cluster to somebody else so it doesn't hit your balance sheet.

[36:27](https://realalpha.app/v/EAtLq9_vrE0/?t=2187) Very simple pitch. Well, thank you so much for your time. You're doing a great service to the industry, and I just am happy to be your friend. I appreciate it. All right. That's it.