Andrew Feldman on Building a Chip 58x Larger Than Nvidia's

[00:00] The demand for AI has outpaced everybody's expectations, everybody's forecasts. And so everybody's chasing. They're chasing chips, they're chasing memory, they're chasing data centers. OpenAI, we announced in January a huge partnership, one of the biggest deals done in Silicon Valley history. We'll be doing more than $20 billion of hardware for them over the next several years. I'm concerned that there are ways frequently for NVIDIA to exercise market strength. [00:30] cheap to limit competition. We have a chance for our children or the next generation not only to not die from cancer but to not know anybody who died from cancer.

[00:49] Andrew Feldman, welcome to Sorcery. Well, thank you so much for having me. We are out here in Paris at Ray's. [00:56] Pretty okay, huh? It's pretty okay. I also heard you're one of the most popular people here. [01:02] That would be first in my life. So it's nice to nice of you to say. [01:08] A big theme, though, is, and we were talking about this just off camera, is hardware is cool. Semiconductors are cool. Everything is in fashion now. [01:17] We are in fashion. We're in demand. We're hard to get.

[01:23] It's a big change. I think what's happened is the demand for AI has sort of outpaced everybody's expectation. [01:30] Everybody's forecast. And so everybody's chasing. They're chasing chips or they're chasing memory or they're chasing data centers. There's an opportunity to invent new things. [01:43] different architectures. It's really a time of sort of a Cambrian explosion of ideas. That's really fun. So you were at Ray's last year. I remember this because I think you had the biggest booth here. There was a fence around it. Yeah. And a lot of people.

What is the biggest difference between last year and this year? Oh, I think last year was their first year. [02:05] And I think there are more people here. [02:11] there, more participants. [02:12] So there are more attendees, there are more companies participating. I think people are participating in a bigger way. [02:19] I think the dinner last night at the secret location called Versailles was sort of awesome. We spend a lot of time in the digital world and the people, we talk about AI, to go look at something that illiterate masons made 400 years ago.

And to sit in it and to enjoy the grandeur is extraordinary. [02:46] This wasn't AI, this wasn't EDA design or CAD, these were guys with pen and paper communicating with masons and other tradesmen who couldn't read. [02:57] And they built something unbelievably beautiful. [02:59] And that was really fun. [03:01] And so you're going to be on stage this year. What are you mainly going to talk about? I'm on stage with Sachin from OpenAI. We announced in January a huge partnership, one of the biggest deals done in Silicon Valley history. We'll be doing more than $20 billion of hardware for them over the next several years.

[03:23] We'll talk a little bit about deploying hardware and fast AI and how important fast inferences in the emerging sort of inference AI landscape. [03:34] Inference is a hot topic. Everyone loves inference. Yeah, we make AI with training. [03:39] And we use AI with inference. And so as the models and as the AI we made becomes useful, [03:47] Everybody wants to use it. [03:49] And inference is the mechanism through which we use it. And so now we have smart AI. People want to use it. And when they want to use it, they want to use it.

And they want to be fast. And that's sort of where we come in. [04:02] And we're the fastest, not by a little bit, but by 20x. And so everybody's using their AI, they're trying new things, they're deploying, you know, GPT or cloud code or one of these supermodels. And [04:20] It's sort of an explosion of [04:23] of building, of trying new things, of a new way to work. [04:27] That's pretty fun. [04:28] So we're nearly two months after your IPO. And there was one post that I thought was really cool.

You posted that you had a very large chip on your shoulder. [04:43] It was quite literal and physical. I hope that, that, [04:48] You know, now that we're running a public company, I wouldn't lose sort of the self-deprecating humor or whatever that I enjoy, the sort of tone in my social posts. And so we put a giant tip. We put this sort of on my shoulder like this and we built a harness for it. [05:10] Yeah, that was a fun one. Yeah. [05:15] I think that... [05:19] people assume that the [05:21] as a CEO or of any size or an entrepreneur that it's sort of always peaches and cream.

[05:27] And it's just not the case. There's an enormous amount of hard work. There's sacrifice that you make and that your family makes. They see you less. It's not for a little bit. It's not like a weekend or two weeks in a row you work hard. It's for years. And so sharing that a little bit is something I thought would be. [05:49] Be well received. [05:50] So what's been the biggest difference for you post IPO? [05:54] - The number of people who want something. - Really? - Yeah, has exploded of one form or another.

[06:01] And we get a little better at triaging those between my executive assistant and chief of staff and just... [06:11] if you had [06:13] 80 emails a day of different people asking you for something. [06:17] That's not work. That's just this range of people who want something of one form or another to meet you, your time, for you to present, for you to donate to their cause, for you to, whoa. That was a little unexpected. [06:32] Your last company, you helped create 100 millionaires. This company... [06:39] You have, I don't know, countless more.

Maybe 1,000. [06:43] Maybe a thousand? Maybe in the future a thousand? Right? [06:48] At IPO, it was about $1,000. [06:51] Would that be investors and employees? As employees. Really? Current and former. Okay. [06:57] Wow. So what lesson did you, [07:00] do you learn from that as a CEO? Like what is in your mind? I think a couple of things I think too. [07:05] to do the job you love to build. [07:08] I think making money is really great and making money for people you care about is really, really great. And when you get a chance to deliver for people who bet on you.

[07:19] who bet chunks of their career. [07:22] You're investors. It's great to deliver for them. They bet on you, but they're diversified. [07:28] They bet on you and 20 other companies. [07:30] When someone bets five or seven years of their career, and a career is 30 years, [07:35] They're betting a sixth of their professional career. [07:39] And when you get to deliver for them and... [07:44] They get to achieve the financial goals that they wanted. That's a great feeling and what I'm proud of every day. [07:52] It's a very selfless...

[07:54] position on that. I mean, it's really interesting because [07:58] Today, I mean, in the backdrop, we are in a hyper super cycle, whatever you want to call it, of AI. And so these companies that are coming up, whether they're in semiconductors or they're in, I don't know, models or coding agents, there's a lot of companies that are rising up really fast and hitting that billion dollar mark. [08:28] How do you maintain a healthy mindset around that? [08:32] I think he... [08:33] You bring it. [08:35] It's not a change, right?

[08:40] um [08:41] and [08:42] for [08:43] the type of people I love working with. They like building stuff and they like building hard stuff when it paid a little, when it was out of fashion. You know, when hardware was uncool, they were still building hardware because that's what they like to build. And now that it's in fashion, they like to build hardware. And they're sort of even keeled about that, that their passion is the building. [09:03] And I think in Silicon Valley, which sort of is what I know, that chasing money is not the path to money.

The path to happiness is working on projects you like with colleagues that are interesting for people with integrity. And if you do that, the money will come. [09:23] But even more importantly, you work on things you like. [09:27] and you'll work with people you learn a little something from and you can teach a little something from. [09:32] And if that's the way you set about pursuing your career, [09:37] I think [09:38] When things are really bad, you're on an even keel. [09:40] And what things are really good, you're uneven keel because you're enjoying what you're doing.

[09:45] This episode is brought to you by Brex, my favorite. You become what you spend on. And I refuse to spend my time on work that shouldn't exist. Expense reports, receipt chasing, and manual closes. The companies building what's next from Vercel, OpenAI, Anthropic, Granola, and DeepGram. [10:05] all made the same call. They all run on Brex. Brex is the intelligent finance platform that combines cards, expenses, and banking into a single stack with agentic finance built in. AI agents that handle expenses automatically [10:20] Enforce policy before spend happens and close your books in minutes.

That's why Sorcery runs on Brex, so I can spend time on building and not busy work. It's time to get Brex AF. Learn more at com slash sorcery. That's b-r-e-x dot com slash s-o-u-r-c-e-r-y. [10:43] Bye. [10:44] So looking forward, I'm sure this is also just a mark in your journey because you're a builder. So what are you most looking forward to over the next couple months, as we said, across this year? Look up. [10:57] Getting to an IPO is not the end of a journey. It's sort of a plateau. It's sort of the arrival at corporate adulthood.

It is the achieving one plateau so that you can climb others. [11:13] And our opportunity has gotten bigger. We have more resources. We're better recognized. [11:21] we can reach more people. [11:23] And we can sort of prosecute our vision and our ambitions with more fuel. And so that's what we're excited about every day. Building more chips, building more data centers, inventing technology that moves the industry forward. [11:41] That's what drives us and gets us out of bed every day. [11:45] I think I was listening to a Harry Stebbings episode that you did and talking about data centers and the demand for them and how we're actually not trying to meet future demand.

We're actually just really behind. We are behind. On where we are. So could you explain where we are with the data center build out? Right. So what happened is data centers had historically moved at sort of the speed of real estate. Right. [12:15] poured concrete and the building would go up and [12:18] Two years ago, [12:20] Nobody cared about AI. All right, two years ago, we were at the beginning of this. And the AI explosions happened so quickly. There's so much demand for computing. [12:30] We've outstripped the demand for compute, for memory.

And where do these things go? They go to these buildings. [12:38] and we haven't made them fast enough. [12:42] And so people are chasing data centers around the around the world. [12:47] And that's what's happening. And so it is a major limitation for for everybody right now. [12:56] We just had Tony Kim on and we were talking about the new architecture of data centers. So how are you working with other companies, whether it's through your build outs, creating the right stack and architecture? Sure. So, you know, the interesting about, you know, data centers hadn't changed a lot in 20 years and a lot of the infrastructure in them hadn't changed.

I mean, the infrastructure. [13:21] The generators that we use for backup have been unchanged for about 20 years. The type of batteries we use for backup, the chillers and the CDUs. And suddenly there's this sort of intense demand and sort of people are looking to innovate. And they're using fuel cells and guys are using jet engines like boom to generate power for these data centers. [13:51] And so there's sort of an innovative push in the data center. For us, while we build this sort of super big chip, the chip that's like 58 times larger than any other chip, it goes in a server.

It goes in a metal enclosure that's about the size of a fridge for a dorm room, a little fridge. [14:21] if you're going to deploy a competitor's product. What Tony's really talking about is sort of these buildings were... [14:31] sort of unimproved for a generation. And now you're part of a factory. [14:37] that makes AI. [14:39] and people are trying to optimize every part of that factory. [14:43] Yeah. And now they're also putting them in space. Yeah. Well, they're talking about putting them in space. I think like a lot of technology, you talk about it for a long time before it happens.

Do you have plans? [14:57] You know, I'm... [15:00] We are really good for space because one of the hardest problems in space is getting all these little chips to talk to each other. And because we're a big chip, we don't have that problem. I don't think we're in danger in the near term of having a data center in space. I think it's more than five years away. [15:19] Five years? Five years. Okay. And that's a long time in our world. [15:24] Right, I mean... [15:26] Three years ago, nobody was using AI.

Right? And so I think we got a lot of work in building data centers on Earth before we actually have big production data centers in space. [15:37] I'd love to talk about how chip design has changed. Co-design has become an important part of how do you create that, [15:47] kind of platform for the next three years because you're building for the next three years. So how do you think about that and how do you go about chip design? Well, I think historically you made chips and you ran software on them.

[16:01] And [16:04] There wasn't surprisingly a close interaction. [16:09] because there was a layer called an operating system that lived between the chip [16:15] software and so you know Intel and AMD [16:21] made chips and [16:23] people wrote software for the operating system or for the chip. [16:28] AI has gotten so large and speed is so important that what they're doing is they're thinking about sort of [16:37] the design together. [16:40] What changes could we make in software that would advantage the hardware? Or as we're designing the hardware, what changes could we make that would make the software easier to run?

And so they're being designed sort of at the same time. [16:52] And like anything, when you sort of design things together, [16:57] Um... [16:58] The advantages are enormous. [17:01] And so this is something that's really taken shape right now. And one of the advantages of our relationship with open AI is [17:09] We get to see exactly where the frontier is going and we get a chance to roll that into our designs. [17:17] One of the advantages Google has is that their TPU can be designed in collaboration with the team building Gemini or the team of DeepMind.

And so they can inform their choices. [17:31] back and forth. And that's an enormously powerful thing that is surprisingly relatively new. [17:38] in our space. [17:39] What do you think the biggest misconception with that process and the challenges are? I think the misconception is that it's easy and all you need to do is get in a room. It's a [17:51] it's a very hard problem. [17:55] You know, the software guys think one way. [17:58] The hardware guys think a slightly different way. [18:02] Anything you do to make it easier to...

[18:05] to write the software makes it harder to do the hardware, right? And these are really hard trade-offs. And so bringing them together and... [18:15] Means these sort of compromises where it will be harder here to make it easier here. [18:21] And that means somebody's schedule is going to be impacted. Somebody's got to add resources. Those discussions are enormously difficult. [18:31] I'm really curious because I come from an outside perspective. You have an inside perspective. There's a lot of big deals that are being thrown around left and right. And I don't know what's actually under the headline.

So when SpaceX comes out and they say they now have multi-billion dollar deals with Google and also with reflection, like what does that actually mean? Like what are they selling them? [18:53] It's tough to tell. [18:55] No, it's tough to tell inside as well. [18:58] Really? Yeah. I think... [19:00] There have been announced some sort of deals that didn't have teeth, [19:05] deals that could have teeth later. I think [19:15] X [19:16] had available capacity. [19:20] And you got to ask why they had available capacity. They had available capacity because the Grok model wasn't used very much.

[19:28] So they had these GPUs that were sitting around and that's a bad idea. [19:33] And so they sold a whole block of them, or released a whole block of them to Anthropic. [19:38] And they looked up and said, whoa, that's a pretty good idea. Right? We had all these GPUs. Our model wasn't a success, but whoa, we can have a great business by sort of stepping into what is a constrained market. Very hard to get lots of GPUs and lease these. And then they looked around and said, well, what else can we, who else can we lease to?

[20:00] And so that's how it started. Now, the specifics of those deals I'm not super familiar with. [20:06] Are you concerned at all about the circular deals that are going on? [20:10] I'm [20:12] I'm concerned that there are ways frequently for NVIDIA to exercise market strength. [20:20] Right. There are ways for NVIDIA to [20:27] Thank you. [20:28] to use their balance sheet. [20:30] to limit competition. [20:33] right, that if they invest in a neocloud, [20:38] right the neocloud is less likely to use a non-NVIDIA chip [20:43] if they invest in a bottle builder.

[20:46] there's pressure not to use other people's ships. [20:51] That's what I'm more worried about. [20:54] Yeah, this is happening in the token world with all the free tokens that are being offered to these startups. That's exactly right. I think these are. [21:04] These are drug pushers. Here's a little girl, try a little bit, just a little bit. And I think what the startups should do is, you know, they should take it and then never be dependent. And take some from AMD and come to us and see if we can get you some as well and avoid dependence.

[21:25] That never ends well. If you're building what's next in AI, you need to know MongoDB, the database platform developers love and built for the agents you're running. MongoDB stores searches and reasons over your data in real time with vector search and embeddings from Voyage AI all in the same system. No separate pipelines, no stitching together 10 different tools. It's why 75% of the Fortune 100 and leading AI native startups run on MongoDB. [21:55] first user to billions of vectors. Go to com slash ai to learn more. That's com slash ai to learn more.

[22:07] Bye. [22:08] Assembly AI is a voice AI infrastructure layer millions of developers build on. They build the industry's best speech-to-text, voice agent, and speech understanding models that serve as critical infrastructure for companies like Granola, HayGen, Ashby, and ClickUp. Their speech-to-text models lead the industry in accuracy and quality, and their speech understanding models help you go beyond transcription by uncovering insights, identifying speakers, and highlighting key information from voice data. [22:38] You can get started today at com/sourcery and get $50 of free credits to start building voice AI products. That's com/sourcery.

[22:51] I mean, this is actually a good transition into something that Karp said not too long ago on CNBC. He was pretty much talking about sovereign AI and how every company should own the full stack. And so whether it's their data for their products and everything in between and models, too, because you don't want to sell that. Then the models will recreate it. And then about a new Figma. How do you feel about sovereign AI in this like the shift over to owning the full stack? [23:21] the notion of not being dependent.

[23:24] I mean, you shouldn't be dependent on NVIDIA. [23:27] you shouldn't be dependent on one model maker. [23:31] I think that [23:34] rarely works out well. [23:36] Thank you. [23:37] What you'd like to be as a nation or as a big company is you'd like to have choices. [23:43] Thank you. [23:44] And I don't know if you need to own all the stack, but you like choices at each layer of the stack. [23:50] I think a different way to think about what Alex said is, [23:55] You want to think about where your advantage is.

[23:58] If you have unique data, [24:03] Be sure you don't give that away, right? Be sure you have... [24:08] multiple choices in different parts of the stack, but you get the credit for your data. [24:13] And it doesn't help make somebody else's model better. [24:17] As we... [24:18] look forward. I guess more on the macro lens, the proliferation of AI and everything that we're able to now create and build and do. What are you excited about on the externalities that come with all of this? [24:33] - Okay. [24:34] Okay, I think.

[24:36] Um... [24:38] that [24:39] We have a chance. [24:41] for [24:42] Our children are the next generation. [24:45] Not only to not die from cancer, but to not know anybody who died from cancer. [24:50] I think that is a real challenge. [24:53] achievable goal. [24:55] in 25 years. [24:58] Wouldn't that be something? [24:59] You know, when we think about what AI can do, [25:03] Writing better code is cool. [25:06] And there's a huge market for that. [25:08] - Thank you. [25:09] But I think what [25:10] what it can do to better humanity.

[25:14] is rid us of the number one killer of adults. [25:18] And I think [25:22] You know, that's when I think of what we're all doing this for. It's an outcome like that. You know, pancreatic cancer had a huge breakthrough recently. I think the opportunity for breakthroughs right now has never been better. And AI is an extraordinary tool in pursuit of knocking down cancer. [25:44] major, major human killers, right? I mean, if you think, if you took out cancer and you took out [25:52] Automobile accidents. [25:55] You're taking out huge numbers of deaths a year.

[25:58] And you say to yourself, "Well, that's a lot of good." [26:02] that we did. [26:04] and I think you know [26:06] self-driving. Humans are terrible drivers. [26:09] Horrible. We're horrible drivers. It's not just when we're 16 or 18 or when we're not paying attention or our parents in their 80s. We don't pay attention. We're on the phone. We're talking to our wife. We're worried about work. That's even before people drink or do all sorts of things that are obviously bad, but we're just not good drivers. [26:30] and machines can drive better than we can today.

[26:33] And so that's the number one killer of people, whatever, 15 to 40 is car accidents. You take that out through self-driving. You take out a major killer like cancer. [26:44] He's going, "Whoa." [26:45] That's a pretty good 20-year run of technology. [26:49] Are you a peptide fan? I'm a peptide fan. You are? Yeah. I think... [26:54] Not in the specifics of there is a peptide, but rather in that our opportunity to advance our knowledge about the biology of our bodies and how to achieve performance and how to achieve longevity and productivity.

[27:12] We're just beginning. Yeah. And we're going to make some big strides, whether it's with peptides or something else or, you know. [27:20] the next GLP-1 inhibitor or whatever. We're going to make giant strides. [27:25] Yeah, it's been really cool to see all the new drug discoveries and new drug discovery companies. Brian Armstrong just came out with this company called New Limit. That's one of them. I think their goal is to eradicate all diseases. It's a good one. Right. I mean, even if that's a little hubris, I mean, it wasn't even hubris that was thinkable a decade ago.

Yeah. Right. I mean, we're now in a realm where, wow, that is... [27:52] sort of crazy big but [27:54] not insane. Right? I mean, how cool is that? Yeah, I mean, well, a lot of people like to talk about the doomerism of AI, but we're I think we're entering a new mix of talking about the actual outcomes with it. And so to your point of what you're talking about, whether it's with AVs or drug discovery, [28:13] It's really good. I think the problem with the doomers is they're only looking at one side of the ledger.

[28:20] I think to look at this with clear eyes, you got to look at both sides of the ledger. You got to say, look. [28:25] We're going to use a lot of power. It's true. [28:27] And AI has some real risks, it's true. And on the other side of the ledger, here's some things it can do really differently. And here's some things it can do that take education. We've known for 2,000 years. [28:41] the right way to educate children, and we never do it. [28:44] I mean, we knew that the right way to educate Alexander the Great was to have a tutor.

[28:50] and the smartest tutor, Aristotle, was, and that you teach each child differently. [28:56] and you think about their way of learning. [28:59] And we never do that. You throw them in a classroom. You teach to some sort of middle level. [29:06] Each child does not get... [29:08] sort of any different teaching for their different way of learning. [29:13] With AI, we can do that. We can get you tutored. It's right for you. [29:16] And what's more, the tutor can be running in the background saying, look, 3% of students make this type of error, and the best way to teach them to overcome this weakness is with this approach.

[29:27] How cool is that? [29:28] I mean, we've been screwing this up for 2000 years. [29:32] And now we can bring it to every child. [29:35] Right? You can put it on the positive side of the ledger. [29:39] And look at clear eyes at both. [29:41] the negative and the positive and see if we're doing right by society. [29:45] So one of our sponsors is Rex, and I bring this up because our questions around performance, they are all about spending smart and moving faster. But performance in terms of like for your finances, I ask this question in terms of your personal side of things.

I do believe that performance for individuals is kind of who you surround yourself with or who you're inspired by or mentored by. [30:08] You've had a great run at building companies and have had great success. I'm really curious who those people are for you. [30:16] There were a couple of mentors. There was a venture capitalist named Pierre Lamont. [30:21] And he's now in his 90s. He invested in us at Cerebris when he was in the ripe age of 84. Oh, my gosh. And was on our board. And he's forgotten more about making chips than I'll ever know.

[30:37] There was a former CEO named Mark Leslie, he was CEO of Veritas. [30:44] they invented the file system [30:46] These were sort of wise people. [30:51] and they had extraordinarily high standards and they [30:57] um [30:59] They taught me a lot about being a leader. [31:01] about demanding a great deal for myself and from others. [31:07] They were people who were exceptional, I think. [31:11] sort of on a day-to-day basis, sort of my co-founders, there are five of us. [31:16] Five is, in almost every case, too many. [31:19] founders.

They all worked, we all worked together in my last company. [31:26] I've learned an enormous amount from them. [31:29] They challenged me and... [31:35] I'm excited to think with them. [31:38] And that's an okay way to go into work every day. [31:42] It's amazing. [31:44] Well, thank you so much, Andrew. It's a pleasure to have time. I know we covered a lot of ground with this conversation. We covered a lot of ground. It's so many topics. I didn't think we were going to get to peptides, I'm going to be honest.

I just had to ask. [31:57] Well, thank you for having me on your show. I really appreciate it. Thank you. [32:00] Huge thank you to the entire RAISE team for an incredible event. And thank you to Brex, MongoDB, and Assembly AI for making this trip and series possible. If you enjoyed this conversation, you're going to love the rest of the RAISE series with Tony Kim from BlackRock, Scott Wu from Cognition, Andrew Feldman from Cerebris, Rodrigo Liang from Salmanova, Michael Hurlston from Lumentum, CJ Desai from MongoDB, and many, many more like our hot takes that we did.

[32:28] at a secret location that you can find on X, YouTube, and Instagram. Subscribe to Sorcery on YouTube for more conversations with the people shaping AI and join the free newsletter, you can also do paid, at vc for weekly insights on AI, robotics, enterprise software, [32:46] consumer, semiconductors, [32:48] Did I say AI? AI again. And everything that's coming next, like funding announcements and all big things in tech. [32:57] Thank you. Bye.