Jensen Huang: NVIDIA - The $4 Trillion Company & the AI Revolution | Lex Fridman Podcast #494

The following is a conversation

with Jensen Huang, CEO

of Nvidia, one of the most important and influential

companies in the history of human civilization.

Nvidia is the engine powering the AI revolution,

and a lot of its success

can be directly attributed

to Jensen's sheer force of will

and his many brilliant bets and decisions

as a leader, engineer, and innovator.

This is the Lex Freedman podcast.

And now, dear friends, here's Jensen Huang.

You've propelled Nvidia into a uh new era in AI,

moving beyond his focus on chip scale design to now rack scale design.

And I think it's fair to say that uh winning for Nvidia for a

long time used to be about building the best GPU possible.

and you still do, but now you've expanded that to extreme

co-design of GPU, CPU,

memory, networking, storage, power, cooling,

software, the rack itself,

the pod that you've announced, and even the data center.

So, let's talk about extreme code design.

What uh is the hardest part of uh co-designing

a system with that many complex components

and design >> Yeah, thanks for that question.

So first of all, the reason why extreme code design is necessary is because

the problem no longer fits inside

one computer to be accelerated by one GPU.

The problem that you're trying to solve is

you would like to go faster

than the number of computers that you add.

So you added, you know, 10,000

computers, but you would like it to go a million times

Then all of a sudden

you have to take the algorithm,

you have to break up the algorithm, you have to refactor it,

you have to shard the pipeline,

you have to shard the data, you have to shard the model.

Now all of a sudden

when you distribute the problem this way,

not just scaling up the problem, but you're distributing the problem,

then everything gets in the way.

This is the AMD doll's law problem

where the amount of speed up you have for something

depends on how much of the total workload it is.

And so if computation

represents 50% of the problem

and I sped up computation

infinitely like a million times

you know I only sped up the total workload by a factor of two.

Now all of a sudden,

not only do you have to distribute the

computation, you have to, you know, shard the pipeline somehow.

Uh you also have to solve the networking problem

because you've got all of these computers are all connected together.

And so distributed computing at the scale that we do,

the CPU is a problem, the GPU is a problem, the networking is a

problem, the switching is a problem,

and distributing the workload across all these computers are a problem.

It's just a massively

complex computer science problem and so we just got to bring every

technology to bear otherwise

we scale up linearly

or we scale up based on

uh the capabilities of Moore's law which has largely slowed because Dernard's scaling has slowed.

I'm sure there's trade-offs there.

Plus you have a completely disperate disciplines here.

I'm sure you have specialists in each one of these high bandwidth memory,

the the networking, the NVL link, the nyx, the the optics and the copper

that you're doing, the power delivery, the cooling, all that.

I mean, there's like world experts in each of those.

How do you get them in a room together to figure out >> That's why

my staff is so large.

>> What's the pro can you take me through the process of the specialists and the generalists?

Like, how do you put together the rack when you know this the set

of things you have to shove into a rack together?

Yeah, >> like what does that process look like of designing it all

>> There's the the first question which is what is extreme code design?

You're we're optimizing across the entire stack of software

from architectures to chips to systems to system software to the algorithms to the applications. That's one layer.

The second thing that you and I just talked about is goes beyond

CPUs and GPUs and networking chips and

scale up switches and scale out switches.

And then of course you got to include

power and cooling and all of that because

you know all these computers are

extremely extremely power power hungry.

They do a lot of work and they're very energy efficient

but they in aggregate

still consume a lot of power.

And so that's one the first question is what is it?

The second question is

why is it and we just spoke about the reason you know you want

to distribute the workload so that you can exceed the benefit

of just increasing the number of computers

and and then the third question is how is it how do you do

it >> and and uh that's the that's kind of the miracle of this company

you know when you're designing a computer you have to have operating system of

computers when you're designing a company

you should first think about what is it that you want the company to produce.

You know, I see a lot of companies organization

charts and they all look the same.

Hamburger organization charts, software organization

charts and car company organization charts.

They all look the same.

And it doesn't make any sense to me.

You know, the goal of au of a company is to be the machinery,

the mechanism, the system that produces

the output and that output is the product that we like to create.

It is also designed

the architecture of the company should reflect

the environment by which it exists.

It almost directly says what you should do with the organization.

My direct staff is 60 people.

You know, I don't have one-on- ones with them because it's impossible.

You can't have you can't have 60 people on your staff

if you're, you know, going to get work done.

And >> so you still have 60 reports.

You still have more. Yeah.

>> And most stars at least have a foot in

almost all of them.

There's experts in memory, there's experts in CPUs,

there's experts in optical all Yeah.

GPUs and architecture, algorithms, design.

>> So, you constantly have an eye on the entire stack

and you're having to do like intense discussions

about the design of the entire stack >> and no conversation

is ever one person.

That's why I don't do one-on- ones.

We present a problem

and all of us attack it,

you know, because we're doing extreme code design

and literally the company is doing extreme code design all the time.

>> So even if you're talking about a particular component

like cooling networking, everybody's listening in.

>> And they can contribute

well this doesn't work for the for the power distribution.

This doesn't exactly >> this doesn't work for the for the memory.

This doesn't work for this. >> Exactly.

And whoever wants to tune out, tune out.

>> You know what I'm saying?

And the reason for that is because because the people who are on the

staff, they they know when to pay attention.

>> They're supposed, you know, something they could have contributed to, they didn't contribute to.

I'm going to call them out, you know, and so, hey, come on, let's get in here.

>> So, as you mentioned, Nvidia is this company that's adapting to the

>> So, at which point can you say,

>> did the environment change?

be began adapting sort of secretly

>> in the early days

from GPU for gaming maybe the early deep learning revolution to we're now going

to start thinking of it as an AI factory.

What does Nvidia do is produces AI.

Let's build a factory that makes AI >> I could I you could I could

reason through it just systematically.

Um we started out as as an accelerator

company but the problem with accelerators

is that the application domain is too narrow.

It has the benefit

of being incredibly optimized for the job.

You know, any specialist has that benefit.

The problem with intense

specialization is that of course your market reach is

but that's that's even fine.

The problem is the market size

also dictates your R&D capacity.

And your R&D capacity

ultimately dictates the influence and impact that you can possibly have in computing.

And so when we first started out in acceler

as an accelerator, very specific accelerator,

we always we always knew that that had that was going to be our first step.

We had to find a way to become accelerated computing.

But the problem is when you become a computing company,

it's too general purpose and it takes away from your specialization.

It turn I connected

two that are actually have fundamental tension.

The better computing company we become,

the worse we become as a specialist.

The more of a specialist,

the less capacity we have to do overall computing.

And so the that and I connected those two words together on purpose

that the company has to find

that really narrow path

step by step by step

to expand our aperture

of computing but not give up on the most important specialization that we had. Okay.

So the first step that we took

beyond acceleration was we invented the programmable pixel shader.

So that was the first step towards programmability.

our you know that was our first journey towards

moving into the world of computing.

The second thing that we did

was we we uh uh created

uh we put FP32 into our shaders.

That FP32 step IE

compatible FP32 was a huge step in the direction of computing.

It was the reason why

um all of the people who were working on

on um stream processors

and you know other types of data flow processors

discovered us and they say hey all of a sudden you know we might

be able to use this GPUs that's incredibly

computationally intensive and it's now

you know compliant with it e I can take my software that I was

writing you know previously on CPUs

and I could you know see about

you know using the GPU for that

and which led us to create put C

on top of FP32

was called we call CG

that CG path took us to eventually

CUDA CUDA step by step by step

um uh we well

putting CUDA on GeForce

that that was a strategic

decision that was very very hard to do

because it cost the company

enormous amounts of our profits and we couldn't afford it at the time

but we did it anyways because we wanted to be a computing

company a computing company

has a computing architecture.

A computing architecture has to be compatible

across all of the chips that we build.

>> Can you can you take me through that decision?

So, putting CUDA on GeForce

could not afford to do.

Can you explain that decision?

Why >> why boldly choose to do that anyway?

>> Can you explain that decision? >> Excellent.

That was that was the first

I would I would say that that was

the first um the first strategic

decision that that is as close to an existential

threat for people who don't know it turned out to be

spoiler alert one of the most

incredibly brilliant decisions ever made by a company.

So CUDA turned out to be

an incredible foundation for computation

uh in this AI infrastructure world.

So, so you're just setting the context.

It turned out to be a good decision.

>> Yeah, it turned out to have been a good decision.

I think the So, so here here's the way it went.

So, we invented this thing called CUDA

and um uh it expanded the the aperture

of applications that that we can accelerate with our accelerator.

The question is how do we

how do we attract developers to CUDA?

Because a computing platform is all about developers.

And developers don't come to a computing platform

just because you know it could perform something interesting.

They come to a computing platform because the install base is large.

Because a developer like anybody else wants to develop software

that reaches a lot of people.

So the install base

is in fact the single most important part of an architecture.

The architecture could attract enormous amounts of criticism.

For example, no architecture

has ever attracted more criticism than the x86.

You know, as as a less than

less than elegant architecture,

but yet it is the defining architecture of today.

It it gives you an example

that in fact so many

risk architectures which were

beautifully architected incredibly well-designed

by some of the brightest

computer scientists in the world

largely failed and so I've given you two examples where

one is you know one is elegant the other one's barely aesthetic

and so yet x86

survived >> install base is everything >> install base defines

an architecture not everything else is secondary. Okay.

And so there were other architectures at the time.

CUDA came out, Open CL was here.

There were you know there's several other competing architectures

but the the thing that the decision that we made that was good was

we said hey look

ultimately it's about um installed base and what is the best way we could

get a new computing architecture into the

By that time frame GeForce had become successful.

We were already selling

millions and millions of GeForce GPUs a year.

And we said, you know, we we ought to put CUDA

on GeForce and put it into

every single PC whether customers use it or not

and use it as a starting point of

cultivating our installed base.

Meanwhile, we'll go and

attract developers and we went to universities

and wrote books and taught classes and put CUDA everywhere.

And eventually people discover and at the time the PC was the primary computing vehicle.

There was no cloud and

we could put a supercomputer

in the hands of every researcher

in school, every scientist,

you know, every engineering school, every or every student in school and eventually

something amazing will happen.

Well, the problem was

CUDA increased our cost

of that GPU, which is a consumer product,

so tremendously it it

completely consumed all of the company's gross profit dollars.

And so, at the time, the company was probably,

you know, worth, I don't know, at the time, eight,

was it like $8 billion

or something like$67 billion

or something like that.

After we launched CUDA,

I recognized that it was going to

add so much cost, but it was something we believed in.

You know, our market cap went down to like $1.5 billion.

And so, we were down we were down there for a while

and and uh we clawed our way

way back slowly, but we carried CUDA on GeForce.

I always say that Nvidia is the house that GeForce

built because it was GeForce

that took CUDA out to everybody.

Researchers, scientists, um they discovered

CUDA on GeForce because they were all, you know, many of them were gamers.

Um many of them built their own PCs anyways

in a university lab.

many of them built clusters themselves

you know using using PC components and

and so that you know that's kind of how we got

>> and then that became the platform the foundation for the deep learning

>> that was also another great great observation

yeah >> that existential moment do you remember

like what were those meetings like

what were those discussions

like deciding as a company

risking everything >> well um I had I had to make it clear to the

board what we were trying to do

and and um uh the management team knew our gross margins were going to get crushed.

So you could imagine

a world where GeForce

would carry the burden

of CUDA and none of the gamers would

appreciate it and none of the gamers would pay for it.

You know, they only pay certain price and it doesn't matter what your cost is.

And so that you know, we we increased our cost by 50%

and that con consumed

and we were a 35% gross margin company.

And so it it was a it was quite a difficult

decision to make, but you could imagine that someday

this could go into workstations

and it would go into supercomputers

and and in those segments maybe we can capture more margin.

Um so you you could you could reason your way into being able to afford this.

Uh but it still took it took a decade.

>> But that but that's more like conversation with the board convincing them.

But you psychologically >> because Nvidia has continued to make

bold bets that predict the future

and in part especially now define the future.

So I'm almost looking for

wisdom about how you were able to make those decisions

to make leaps like that as a company.

Well, f first of all,

um I'm informed by

by by a lot of curiosity.

Uh at some point

there's a reasoning system

that that convinces me

uh so clearly this outcome will happen

that this will happen.

And so I believe I believe it in my mind.

And when I believe it in my mind,

you know, you know how it is.

You manifest a future.

And that future is

so convincing, there's no way it won't happen.

There's a lot of suffering in in between,

but you've got to believe what you believe.

>> So you you you envision the future.

>> And you essentially from a sort of engineering perspective manifest it. >> Yeah.

And and you you reason about how to get there.

You reason about why it it must exist.

Um and and um and you know, I reason

we all reason here.

the management team will reason about it.

All the people that I we spend a lot of time reasoning about it.

The thing the thing that

the next part of it is probably a skill thing which is

you know oftentimes in leadership

uh the leadership stays quiet or they learn about something and then they do

some manifesto and it's a brand new year and somehow at the end of

the year next year we're going to have a brand new plan,

big huge layoff this way, big huge organization

change this way, new mission statement,

brand new logos, um you know that kind of stuff.

Um, we've just never

I never do things that way.

When I learn about something

and it's starting to influence how I think, I'll make it very clear to

everybody near me that, you know, this this is interesting.

Um, this is going to make a difference.

Uh, this is going to impact that.

And I reason about things step by step by step.

often times I've already made up my mind

but I'll take every possible opportunity

external information new insights

new discoveries uh new engineering

you know revelations uh new milestones

developed I'll take those opportunities

and I'll use it

to shape everybody else's

belief system and I'm doing that literally

every single day I'm doing that with my board

I'm doing that with my management team I'm doing that with my employees

I'm trying to shape

their belief system such that when I come the day I say,

"Hey, let's buy Melanox."

It's completely obvious to everybody

that we absolutely should.

On the day that on the day that I that I said, "Hey guys,

let's go all in on deep learning."

And let me tell you why.

I've already been laying down the bricks

to different organizations inside the company.

every organization and every everybody

many of the people might have heard everything

most of the company heard hears of course pieces of it

and on the day that I announce it

um everybody's kind of bought into many pieces of it

and in a lot of ways I like to announce these things

and I imagine um that that

the employees are kind of saying you know Jensen what took you so long

and and in fact I've been shaping their belief system for some time and

therefore leadership sometimes it looks like you're leading from behind

>> but you've been shaping their you know to the point where on the day

that I declared it

100% buy in but that's what you want you want to bring everybody along

you know otherwise we announce something about deep learning and everybody goes what are

you talking about you know you announce

something about let's go allin on this thing

and and your your management team your board your employees your customers,

they're kind of like, where's this coming from?

You know, this is insane.

And so, so, uh,

GTC, in fact, if you go back in time, you look at look at

the keynotes, I'm also shaping

the belief system of my partners and the industry and and I'm using that

to shape, you know, the belief system of my own employees

and and and so by the time that I announce something,

like, for example, we just now we just announced Grock, we've been late.

I've been talking about the

stepping stones for two and a half years.

You guys just go back and

oh my gosh, they've been talking about it for two and a half years.

And so I've been laying the foundation step by step by step.

So when the time comes you announce it, everybody's,

you know, what took you so long?

>> But it's not just inside the company.

You're shaping the landscape,

the broader global landscape of innovation.

Like putting those ideas out there, you really are manifesting reality.

>> We don't build computers.

We actually don't build clouds.

We don't, as it turns out, we're a computing platform company

and so nobody can buy anything from us.

That's the weird thing.

You know, we ver we vertically

design vertically integrate to design and optimize,

but then we open up the entire platform

at every single layer

to be integrated into

other companies products and services and clouds and

supercomputers and OEM computers and

and so the amazing thing is

I can't do what I do

without having convinced them first.

And so most of GTC

is about manifesting a future

that by the time that we

my product is ready,

they're going what took you so long? Yeah.

Uh so one of the things you've been a believer

for a long time

is uh scaling laws broadly defined.

So are you still a believer in the in the scaling

>> Yeah, we have more scaling laws now.

>> So I think uh you've outlined

four of them with pre-training,

post- training, test time, and agentic scaling.

What do you think

when you think about the future,

deep future and the near-term

future, what are the blockers

that you're most concerned about that keep you up at night that you have

to overcome in order to keep scaling?

>> Well, we can go back and reflect on what people thought were blockers. >> Mhm.

So in the beginning we were the first the pre

pre-training scaling law you know people thought

uh well rightfully so that the amount of data that we have

high quality data that we have

um will limit the intelligence that we achieve and that scaling law was an

important very important scale law the larger the model

the correspondently more data

uh results in a better with a results in a smarter AI

and so that was pre-training

and Ilas Susker Ilas

we're out of data or something like that.

Pre-training is over or something like that.

The the industry panicked,

you know, that this is the end of AI.

And of course, of course, that's that's obviously not true.

Um, we're going to keep on scaling the amount of data that we h

have to to train with.

A lot of that data is probably going to be synthetic.

And that also confused people, you know, and and what people don't realize is

they've kind of forgotten

that most of the data that that we are training

uh that we teach each other with, inform each other with this is synthetic.

You know, I it's synthetic

because it didn't come out of nature. You created it. I'm consuming it.

I modify it, augment it,

I regenerate it, somebody else consumes it.

And so so we've now reached a level

where AI is able to

take ground truth, augment it,

enhance it, synthetically generate an enormous amount of data

and that part of post training

um continues to scale.

And so the amount of data that we could use

that is human generated

will be smaller and smaller and smaller.

the amount of data that we use to

uh train model uh

uh is going to continue to scale

to the point where

we're no longer limited

training is no longer limited by data is now limited by compute

and the reason for that is most of the data is synthetic

then the next phase

is uh test time

and um I I still remember people people telling me that inference

oh yeah that's easy

pre pre-training that's hard these are giant systems that people are talking about

inference must be easy

and so inference chips are going to be little tiny chips and

you know they're not they're not like Nvidia's

chips oh those are going to be

complicated and expensive and you know we could make and this is and in

the future inference is going to be the biggest market and it's going to

be easy and we're going to commoditize

and you know everybody can build their own chips and

and and that was always

illogical to me because

inference is thinking and I think thinking is hard

thinking is way harder than reading.

>> You know, pre-training is just

memorization and generalization, you know, and looking for patterns and relationships.

You're reading and reading

versus thinking, reasoning, solving problems,

taking un unexplored experiences,

new experiences, and breaking it down into

de decomposing it into,

you know, solvable pieces

that we then go off either through first principal reasoning or,

you know, through through uh previous examples,

prior experiences, you know, or or or

just uh uh exploration.

and and search and

you know trying different things and

that whole process of

post of of test time scaling.

Uh inference is really about thinking

and and it's about reasoning. It's about planning. It's about search.

It's about and so how could that possibly be computed?

And we were absolutely right about that you know so so test time scaling

is intensely comput intensive.

Then the question is okay now we're at inference and we're at test time scaling. What's beyond that?

Well, uh we have now created

you know one agentic

person and that one agentic

person has a large language model that we've now we've now you know developed.

But during test time, that agentic

system goes off and does research

and bangs on databases

and it goes on and you know uses tools and one of the most

important things it does is spins off and spawns off a whole bunch of

sub aents which means we're now creating large teams.

It's so much easier

to scale Nvidia by hiring more employees

than it is to scale myself.

>> And so the next scaling law is the agentic scaling law.

It's kind of like multip multiplying AI.

Multiplying AI, we could spin off agents as fast as you want to spin off agents.

And so, you know, I you have four scaling laws.

And and as we use the a agentic

systems, they're going to create a lot more data.

They're going to create a lot of experiences.

Some of it we're going to say, "Wow, this is really good.

We ought to memorize this."

>> That data set then comes all the way back to pre-training.

We memorize and generalize it.

We then refine it and fine-tune it

back into post training.

Then we enhance it even more with test time, you know, in the agent

agents agentic systems, you know, put it onto the indust industry.

And so this loop,

the cycle is going to go on and on and on.

It kind of comes down to basically

intelligence is going to scale

by one thing and it's compute.

But there's a tricky thing there that you have to anticipate and predict

which is some of these components.

It requires different kind of hardware

to really do it optimally.

So you have to anticipate

where the AI innovation is going to lead.

For example, mixture of experts with sparity.

>> With hardware, you can't just pivot on a week's notice.

You have to anticipate what that's going to look like.

That's >> that's so scary and difficult to do, right?

For example, uh these AI model architectures

are being invented about once every six months. >> Yeah. Right.

And uh system architectures

and hardware architectures kind of every 3 years.

And so you need to anticipate

what likely is going to happen,

you know, 2 3 years from now.

And there's a couple ways that you could do that.

First of all, we could do research internally ourselves.

And that's one of the reasons why we have basic research.

We have applied research.

We create our own models.

And so we have we have hands-on

life experience right here.

This is part of the code design that I'm talking about.

>> We're also the only AI company in the world that works with literally every

AI company in the world.

And to the extent that we can

um uh we try to get a sense of of what are the challenges

that people are experiencing.

>> So you're listening to the whispers

across the industry, the adabs. >> That's right.

You got to listen and and learn from everybody and have a have a

and then the the last part is to have an architecture

that's that's flexible that can adapt

and move with the wind and one of the benefits of of CUDA is

that it's you know on the one hand

an incredible accelerator on the other hand

it's really flexible and so that balance

incredible balance between otherwise we can't accelerate the the CPU

versus generalization so that we can adapt with changing algorithms.

That's really really important.

That's the reason why

why um CUDA has been so resilient

um on the one hand

and yet we continue to enhance it.

We're at CUDA 13.2

and so we're invol evolving

the architecture so fast

that we can stay with

you know with with

the modern al algorithms.

Um for example uh when mixture of experts came out

uh that's the reason why we had MVLink

72 instead of MVLink 8.

We could now take an entire

4 trillion 10 trillion parameter model and put it in one computing domain

as if it's running on one GPU.

Um I people probably

didn't notice I said it

but if you look at the architecture

of the Grace Blackwell

racks it was completely focused on doing one thing processing the LLM.

All of a sudden one year later

you're looking at a Vera Rubin rack.

It has storage accelerators.

It has this incredible new CPU called Vera.

It has Vera Rubin

and MVLink72 to run the LLMs.

It also has this new additional rack called Gro.

And so this entire rack system

is completely different than the previous one

and it's got all these new components in it.

And the reason for that is because the last one was designed to run

large language models inference

and this one is to run agents

and agents bang on tools

and Obviously the design of the

had to have been done

before claude code, codeex, open claw.

So you were anticipating

the future essentially and that that comes from what?

From the whispers, from the understanding what all the state of the artist is.

>> No, it's it's easier than that.

Uh you you just reason about it.

Uh first of all

just reason no matter

no matter what happens at some point

in order for that large language model to be a digital worker.

Let's just let's just use that metaphor.

Let's say that we want the LM to be a digital worker.

What does it have to do?

It has to access ground truth.

That's our file system.

It has to be able to do research.

It doesn't know everything.

We don't have and I don't want to wait until this AI

becomes, you know, universally

smart about everything past,

present, and future before I make it useful.

And so therefore, I might as well let it go do research.

It's obviously if it wants to help me, it's got to use my tools.

You know, a lot of people would say,

you know, um AI is going to completely destroy software.

We don't need software anymore.

We don't even need tools anymore. That's ridiculous.

Let's let's use the

let's use a thought experiment.

Uh, and you could just sit there, enjoy a glass of whiskey

and and think about all these things and it would become completely obvious

like if I were to

create the most amazing

ro the most amazing

agent that we can imagine in the next 10 years, let's say be a humanoid robot.

If that human or robot were to be created,

is it more likely that the human or robot comes into my house

and uses the tools that I have

to do the work that it needs to do?

Or does his hand turns into a

10- pound hammer in one instance,

turns into a scalpel

in another instance, and in order to boil water,

it beams, you know, microwaves

out of its fingers,

you know, or is it more likely just to use the microwave,

you know, and the first time it goes up to the microwave.

It probably doesn't know how to use it. But that's okay.

It's connected to the internet.

It reads the manual

of this microwave, reads it

instantly, becomes an expert, and so uses it.

>> And so I I think the

I just described in fact

almost all of the

properties of Open Claw.

>> You know, that it's going to use tools, that it's going to access files,

it's going to be able to do research,

it has IO subsystem.

And when you're done reasoning through it, reasoning about it through through it in

that way, um then you say, "Oh my gosh,

the impact to the future computing is deeply profound."

And the reason for that is I think we've just reinvented the computer.

And then now you say, "Okay,

when did we reason about that?

When did we reason about Open Claw?"

If you take the Open Claw

schematic that I used at GTC,

you will find it two years ago.

Literally two years ago at GTC,

I was talking about

Asgentic systems that exactly

reflect open claw today

and and of course the confluence

of of many things had to happen.

First of all, we needed

claude and and GPT

and you know all of these models

to reach a level of capability.

So so their innovation and their breakthroughs and their continual advances was really important.

And then of course somebody had to create a an

open- source you know

um project that that

uh was sufficiently robust

you know and sufficiently

complete and that we can all we can all put to put to work

and and I think openclaw

did for did for agentic

systems what chat GPT did for generative systems and and I just think it's

a very big deal.

>> Yeah, it's a really special moment.

I'm not exactly sure why it captured

so much of the world's attention, but it did more than cloud code and

codeex and so on

because consumers could reach it. >> Sure. Yeah.

But there there's also

so much of this is vibes and

and Peter uh I had a podcast with him.

He's a wonderful human being.

So part of it is also the humans that represent the thing.

Part of it is memes and the

>> cuz we're all trying to figure it out.

There's really serious and complicated security concerns about

when you have such powerful technology, how do you hand over your data so

they can do useful stuff, but then there's scary things associated with that.

And we as a civilization,

as individual people and as a civilization

figuring out how to find that right balance.

>> Yeah, we we uh we jumped on it right away and we sent a

bunch of security experts this way

>> and we did this thing called Open Shell.

It's it's already been

integrated into into open claw

>> and Nvidia put forward Nemo claw. >> Yep. Exactly.

>> The install is super easy.

It makes sure that uh it's secure.

>> We give you two out of three rights.

Agentic systems can can access sensitive information.

It can execute code and it can communicate externally.

>> We could keep things safe if we gave you two out of those three

capabilities at any time, but not all three.

And out of those two out of three capabilities,

we also give you access control based on based on um whatever rights that

you're given by enterprise.

And then we connected to a policy engine

that all these enterprises already have.

And so um we're going to try to do our best to to

uh help Open Claw become a a better claw.

So you eloquently explained

how we have a long history of blockers that we thought were going to be

blockers and we overcame them.

But now looking into the future, what do you think might be the blockers

now that it's clear that agents will be everywhere?

So it's obviously we're going to need compute.

So what is going to be the blocker for that scaling?

Power is a concern, but it's not the only concern.

But that's the reason why we're pushing so hard on extreme code design

so that we can

improve the tokens per second

per watt orders of magnitude every single year.

And so in the last 10 years,

Moors law would have

progressed computing about a 100 times in the last 10 years.

We progressed and scaled up computing by a million times in the last 10 years.

And so we're going to keep on we're going to keep on doing that

through extreme code design.

Um so energy efficiency

per per watt completely

affects the revenues of a company.

It affects the revenues of a factory

and we're just we're just going to push that to the limit so that

we can keep on driving token cost down

as fast as we can.

you know, the our computer

price is going up,

but our token generation

effectiveness is going up so much faster

that token cost is coming down.

It's just it it's coming down an order of magnitude every year.

>> So power that's an interesting one.

So the the way to try to

get around the power blocker is to try to with the tokens per second

per watt try to make it more and more efficient.

Of course, there's the question, how do we get more

>> We should also get more power.

>> That's a really complicated one.

And you've talked about small module nuclear power plants.

There's all kinds of ideas for energy.

Uh how much does it keep you up at night?

Uh the the bottlenecks

in the supply chain of AI

like ASML with EUV lithography

machines, TSMC with advanced packaging

like cos and uh SK HX with

high bandwidth memory all all the time and we're working on all the time.

No company in history

has ever grown at a scale that we're growing

while accelerating that growth.

It's >> And it's hard for people to even understand this

in the overall world of AI computing. We're increasing share.

And so supply chain upstream

and downstream are really important to us.

I spent a lot of time

um informing all the CEOs that I work with

what are the dynamics

that's going to cause

uh the growth to continue or even accelerate.

It's part of the reasons why

to the entire right hand side of me

were CEOs of practically the entire IT industry

upstream and practically the entire infrastructure industry downstream. Mhm.

And they were all

there were several hundred CEOs

and I don't think there's ever been keynotes where several hundred CEOs show up.

And and part of it is I'm telling them about our business condition now.

I'm telling them about

the growth drivers in the very near future and what's happening.

And I'm also describing

where are we going to go next

so that they could use all of this information

and all of the dynamics that are here

to inform how they want to

And so so I

I inform them that way like I inform my own employees.

And then of course then I make trips out to them

and make sure that hey listen I want you to know

this quarter, this coming year, this next year

these things are going to happen

and and if you look at the CEOs of the DRAM industry

um the number one DRAM

in the in the world was

DDR memory for CPUs in data centers.

About three years ago,

I was able to convince several of the CEOs

that even though at the time HBM memory was used quite scarcely,

you know, and and barely by

um that this was going to be a mainstream

memory for data centers in the future.

And at first it sounded ridiculous,

but several of the CEOs believed me and decided to invest

in building HBM memories.

Another memory was rather odd to put into a data center

is the low power memories that we use for cell phones.

And we wanted them to adapt them

for supercomputers in the data center.

And they go, cell phone memory for supercomputers.

And I explained to them why.

Well, look at these two memories, LPDDR5,

The volumes are so incredible.

All three of them had record years in history.

And these are these are 45 year old companies.

And so, you know, I that's part of my job is to

inform and shape, you know.

So, you're not just manifesting

the the future and maybe inspiring

Nvidia, the the the different engineers of the company.

You're you're manifesting the supply chain of the future.

So you're having conversations with TSMC,

with ASML, >> upstream, downstream, >> upstream, downstream.

So that's the >> GEV, Caterpillar.

>> Yeah, that's downstream from us. Yeah. Yeah. There you go. >> Yeah. The whole thing.

I mean, but that's so

>> there's so much incredibly

difficult engineering that happens

in the the entire semiconductor industry.

And it's just feels

scary how intricate the supply chain is, how many components there are, but it works somehow. Exactly.

The deep science, the deep engineering,

the incredible manufacturing, and so much of the manufacturing is already robotics,

but we have a couple of hundred suppliers

that contribute the technology that goes into our 1.3 million component rack.

>> Each rack is 1.3

one and a half million components.

There are 200 suppliers

across the Vera Rubin rack.

>> So, it's interesting that you don't list that as the thing that keeps you

up at night in the list of blockers.

>> But I'm doing I'm doing all the things necessary to

>> See, I can go to sleep because I checked it off.

I said, "Okay, you know, I I go I I can go to sleep

and I go, well, let's see what

um re let's reason about this.

What's important for us?"

Um because okay let's reason about this

uh because we changed the system architecture

from the original DGX1

that you remembered to uh MVLink

72 rack scale computing.

>> What's going to what does that what does that mean?

What does that mean to uh software?

What does that mean to engineering?

What does that mean

uh to how we design and test and what does that mean to the supply chain?

Well, one of the things that it meant

was we moved um supercomput

superco computer integration at the data center

into supercomputer manufacturing in the supply chain.

If you're doing that,

you also have to recognize you're going to move one

and and if if if

you're if you're, you know, total

footprint of whatever data center you're going to build, let's say you would like

to have, you know,

50 gawatts of supercomputers

that are running and it takes one week to manufacture that 50 gawatts of supercomputers.

Then each week in the supply chain,

the supercomputers are going to need a gigawatt of power.

And so so we're going to need the supply chain to increase the amount

of power it has to

build test to build and test the supercomputers

in the supply chain before I ship it.

>> Well, MVLink72 literally builds supercomputers

in the supply chain

and ships them two, three tons at a time per rack.

It used to be come they used to come in parts and we used

to assemble them inside the data center.

But that's impossible now because MVLink

72 is so dense.

And so that's an example.

And I would have to go

into, you know, I fly into the supply chain, go meet my partners, and

hey, I said, guess what?

So here's what we're going to do with

this is the way we used to build our DGXs.

We're going to build them this way.

This is going to be so much better because we're going to need them for inference.

The market for inference is, you know, coming.

The inflection point for inference is coming.

It's going to be a big market.

And so I first explain to them what's going on, why it's going to

happen, and then I then I

ask them to make several

billion dollars of capital investments

each and because they,

you know, they trust me and and I I I'm very respectful of them

and I I give them every opportunity to question me and I spend time

to explain things to people and I reason about it.

I draw them pictures and I reason about it in first principles

and by by the time I'm done with them there's no what to do.

>> So it's a lot of is about relationships

and building a shared

view of the >> Uh but do you worry about certain bottlenecks?

I mean what are the biggest bottlenecks in the supply chain?

Are are you worried about it ASML V tooling?

Are you are you worried about

the the packaging co-as packaging of TSMC

about how fast it could scale?

like you said, you're

not only growing incredibly fast, you're accelerating a growth.

So it it it feels like

every everybody in the supply chain and those are certainly bottlenecks

would have to scale up.

>> Are you having conversations

with them like how can you scale up faster?

>> Do you worry about it?

>> Because because I told them what I needed,

they understood what I need.

They told me what they're going to go do and I believe in what

they're going to >> Interesting.

That's great to hear.

So maybe if we can just linger on the power for a little bit.

Uh what are your hopes for

how to solve the energy problem?

One of the areas

le that I'm um

that I would love I would love

love us to talk about and just get the message out.

You know um our

our our power grid

is designed for the worst case condition

with some Well, 99%

of the time we're nowhere near the worst case condition because the worst case

condition is a few days in the winter, a few days in the summer and extreme weather.

Most of the time

we're nowhere near the worst case condition and we're probably running around call it 60% of peak.

And so 99% of the time

our power grid has excess power

and they're just sitting idle.

But they have to be there sitting idle because just in case when the

time comes hospitals have to be powered and you know infrastructure

has to be powered and airports have to run and so on so forth.

And so the question that I have is

whether we could go and

um help them understand

and create contractual agreements

and design computer architecture

systems, data centers such that

when they need um the maximum power for

infrastructure in society that the data centers would get less.

>> But that's in a very rare instance anyways.

And during that time, we either have our backup generator for that little part

of it or we just have our computers

shift the workload somewhere else or we have the computers just run slower.

You know, we could degrade our performance,

reduce our power consumption

and provide for, you know, slightly longer latency

response, you know, when somebody asks for, you know, asked for an answer.

And so I think that that

that way of using computers

of building data centers

instead of expecting 100% uptime

and these contracts that are really

really quite rigorous it's putting a lot of pressure on the grid

to be able to now they're going to have to

increase from their maximum.

I just want to use their excess.

It's just sitting there. Yeah.

That's not talked about enough.

So what's what's this what's stopping there? Is it regulation?

Is it >> I think it's it's a throughway problem.

Uh it starts with the end customer.

The end customer puts

puts requirements on the data centers

that they can never not be available. Okay.

So that the end customer expects perfection.

Now in order to deliver that perfection,

you need a combination of

backup generators and your

grid power supplier to deliver on perfection.

And so everybody's got to have 69s.

>> Well, I think first of all,

right now, we ought to have everybody understand that when the customer asks for

these things, you got somebody, you have somebody in your data center operations

team disconnected from the CEO.

I bet the CEO doesn't know this.

I'm going to talk to all the CEOs.

The CEOs are probably

not paying any attention to the contracts that are being signed.

And so everybody wants to sign the best contract of course and they go

down to the cloud service providers and the contract

the the two contract negotiators

that are you I could just see them now

>> you know negotiating these multi-year

contracts both sides want

you know the best contract

as a result the CSPs

then have to go down to the utilities and they expect the nine the

69s and so I think I think the first thing is just make sure

that that all of the customers,

the CEOs of the customers realize what they're asking for.

Now, the second thing is we have to build data centers that gracefully degrade.

And so, if the power, if the utility of the grid tells us,

listen, we're going to have to back you down to about 80%.

We're going to say that's no problem at

>> We're just going to move our workload around.

We're going to make sure that

data is never lost,

but we can reduce the

computing rate and use less energy.

the quality of service degrades a little bit

for the critical workloads

I shift that somewhere else right away

so I don't have that problem

and so you know whoever whichever

data center still has 100% uptime and so

how difficult of an engineering problem is that the smart dynamic allocation of power

in the data >> as soon as you could specify

you could engineer it

beautifully put so long as it obeys the laws of physics on first principles

I think we're good

>> what was the third thing you were mentioning

um so the Second thing is the the data centers

>> and the third thing is we need the

to also recognize that this is an

>> and and instead of instead of saying look

um it's going to take me 5 years to increase

my grid capability uh if you if you have if you're willing to take

power of this level of

guarantee I can make them available for you next month

and at this price

and so if utilities He's also offered

more segments of power delivery promises,

then I think everybody will figure out what to do with it. Yeah.

But there's just way too much waste in the in the grid right now.

We we should go after

>> Uh you've uh highly lauded Elon

and uh Xi's accomplishment

in Memphis in building

um Colossus Supercomputer probably in record time in just 4 months.

It's now at 200,000

GPUs and growing very quickly.

Is there something that you could speak to the

understand about his approach

that's instructive to the broadly to all the data center creators

that's um that enabled that kind of accomplishment

his approach to engineering his approach to the whole

management of construction everything

first of all Elon is

deep in so many different topics

um uh yet he's also

a really good systems thinker

>> and so he's able to think through multiple disciplines

and and um uh he

uh pushes things questions

everything whether number one

is it necessary number two

does it have to be done this way and number you know does it

have does it have to take this long

and and so so he

he has he has the he has the ability

uh to question everything

uh to the point where

everything is down to its minimal amount that's necess necessary.

You can't take anything else out

and and yet yet the the

uh the the the the

necessary um capabilities of the product retains,

you know, and so he's he is as minimalist

as you could possibly imagine

and he does it at a system system scale.

Um I I also love the fact that he he is um

he is represented he he is he is present

at the point of action. >> Mhm.

you know, he'll just go there

and if there's a problem, he'll just go there

and show me the problem.

You know, when you do all of this in combination,

you overcome a lot of previous

this is just the way we do it.

>> Um, you know, I'm I'm waiting for them.

I, you know, I mean, just everybody has a lot of excuses.

And and so and then and then the last thing is

when when you act personally with so much urgency,

uh it causes everybody else to act with urgency,

you know, and and every supplier has a lot of customers going on.

Every supplier has a lot of projects going on.

And he he make it he made it he makes it his business

that he's the top priority of everybody else's, you know, projects.

And so he does that by demonstrating it. >> Yeah.

I've been in a bunch of those meetings.

is it's fun to watch cuz really

not enough people ask the question like

okay so uh can this be done a lot faster and how

why does it have to take this long yeah >> and then that becomes an

engineering question often and yes I think when you get the ground truth of

actually I remember um

one of the times I was hanging out with him he literally is going

through the entire process how to plug in cables into a rack

and he's was working with engineer

on the ground that's doing that task and he's just trying to understand what

does that process look like so it can be less

and just building up that intuition from every single task

involved in uh putting together the data center.

You start to immediately

get a sense at the

detailed scale and at the broad system scale

of where the inefficiencies

are and so you can make it more and more and more efficient.

Plus, you have the big hammer of being able to say, "Let's do it totally different."

>> And remove all possible blockers. >> That's right.

>> Is there parallels in the Nvidia extreme systems code design approach that you see

in the way Elon approaches systems

>> Well, first of all, the code design is a ultimate systems engineering problem.

And so, we approach we approach the work that we do from that first from that principle.

Um the other thing that we do

uh and this is this is a a philosophy

that a thought a a state of mind I guess

a method that I started

uh 30 years ago

and it's called the speed of light.

The speed of light is not just about the speed.

Speed of light is my my

shorthand for u what's

what's the limit of what physics can do.

And so every single everything everything that we do is compared against the speed of light.

Um memory speed uh math speed

uh power cost time

effort number of people manufacturing cycle time.

And uh when you think about latency versus throughput,

uh when you think about cost

versus throughput, cost versus

capacity, all of these things,

uh you test against the speed of light

to achieve all of these different constraints separately.

And then when you consider it together,

you know, you have to make compromises

because a system that achieves extremely low latency

versus achie a system that achieves very high throughput

are architected fundamentally differently.

But you want to know what's the speed of light of a system

that achieves high throughput?

What's the speed of light of a system that achieves low latency?

And then when you think about the total system,

you can make trade-offs.

And so I I force everybody to think about what's this what the f

the first principles the limits

>> the physical limits um for everything before we

you know before we uh do anything

and and we test everything against that and so that's a good frame of

mind I don't love

the other methods which is continuous

>> the the problem with continuous improvement

it it first of all you should engineer

something from first principles at the speed you know with speed of light thinking

limited only by physical limits

and and physics limits

and um after that of course you would

improve it over time

um but I don't like going into a problem and somebody says hey you

know it takes 74

days to do this

>> right now and um we can do it for you in 72 days

>> you know I rather strip it all back to

>> and so first of all explain to me why it's 74

is in the first place

and let's know let's think about what's possible

today and if I were to to build it completely from scratch

you know how long would it take often times you'd be surprised and might

come to 6 days

now the rest of the 6 days to 74

could be very wellreasoned

and compromises and you know cost reductions

and all kinds of different things

but at least you know what they are

and then now that you know that six days possible

Then the conversation from 74 to 6

surprisingly much more effective

>> in such incredibly complex systems that you're working with is simplicity

sometimes a good huristic

to to reach for

I mean if I can just

I mean the pod

the Vera Rubin pod that you announced is just incredible

uh we're talking about seven chips

seven chip types five purpose-built

rack types 40 racks 1.2 two quadrillion

transistors, nearly 20,000 Nvidia dies,

over 1100 Ruben GPUs,

60 exoflops, 10 pabytes per second of scale bandwidth.

Uh, that's all just one

>> that's just one pod.

>> That's just >> Yeah, that's just one pod.

>> I mean, so you have the and then even the the NVL72

rack alone is 1.3

million components, 1300 chips, 4,000 lb crammed into a single 19inch wide rack.

And Lex, we'll probably kind of crank out about 200 of these pods

a week just to put in

>> the the amount of different components.

I suppose simplicity is impossible,

but is that a metric that you kind of

reach for in trying to design things?

>> You know, the phrase the phrase that I use most often is

we we need things to be as complex as necessary but as simple as possible.

And and so the question is is all that complexity there necessary?

And we ought to test for that and we ought to challenge that.

And then after that everything else above it,

you know, it's gratuitous.

>> But it's some of the most incredible

semiconductor industry broadly, but what Nvidia is doing

uh some of the greatest engineering in history.

So these systems are just truly truly marvels of engineering.

>> It is the most complex computer the world has ever made.

Yeah, the engineering teams.

I mean, I don't it's not a competition, but I don't know if if

it was like an Olympics of uh engineering teams.

I mean, TSMC does incredible engineering.

Like I said, ASML

at every scale, but Nvidia

is going to give them a run for their money.

>> Just incredible, incredible teams,

>> gold medal medalist in every single

in every single sport, all assembled right >> and have to work together

and report directly to you. This is wonderful.

Uh you've recently traveled to China.

Uh so it's interesting to ask you

uh China's been incredibly

successful in building up its technology sector.

What do you understand about

um how China is able to over the past 10 years build so many

incredible world-class companies, world-class

engineering teams and just this technology ecosystem

>> that produces so many um incredible products.

whole bunch of reasons for well first of all let's let's start let's start

with some facts 50% of the world's AI researchers are Chinese

plus or minus and

they're mostly in China

still we have many of them here

but there's amazing researchers still in China

um they their tech industry

showed up at precisely the right time

at the time of the mobile cloud era

uh their way of contributing was software and So this is a country's

in incredible science and math.

Uh really well educated kids.

Um uh their tech industry was created during the era of software.

They're very comfortable with modern

China is not one giant economic country.

It's got many provinces and cities

with mayors all competing with each other.

That's the reason why there's so many EV companies.

That's the reason why there's so many AI companies.

That's the reason why there's so many

every company you could imagine.

Um they all create

some of them and

and um as a result

they have insane competition

internally and you know what remains

is an incredible company.

Um they also have a um social culture

where where it's family first,

friends second and company third.

And so the amount of

conversation that goes back and forth

between they're essentially open source all the time.

So the fact that they contribute more to open source

is so sensible because

they're probably what are we protecting?

You know my engineers

their brothers are in that company their friends are in that company and they're all schoolmates.

you know the schoolmate concept

it's a you know one schoolmate

your brother for life

and um and so they

they they share knowledge very very quickly

and so there's no sense

keeping technology hidden you might as well put it on open source and so

the open source community then amplifies

accelerates the the innovation

process so you get this rapid

incredibly great talent rapid

innovation because of open source and just you the the nature of friends

and and um insane

competition among compet among the company what emerges

is incredible stuff and so

this is the fastest innovating

country in the world today and this is something that has everything that everything

that I've just said

is fundamental to just how the kids were grown

the fact that they have excellent education

the fact that they

parents want them to do well in school

the fact that they their culture that way.

These are, you know, these are just the thing about their country

and they showed up at a precisely

the time when technology

is going through that exponential.

>> Plus, culturally, it's pretty cool to be an engineer.

It connects to all the components that you're mentioning.

>> It's a it's a builder nation.

>> It's a builder nation.

>> Yeah, it's a builder nation.

Um, our country's leaders, incredible,

but they're mostly lawyers.

They're country's leaders and because we're they're trying to keep us safe.

uh rule of law, uh governing.

Their country was built out of poverty

and so most of their leaders are

incredible engineers, some of the brightest minds.

To take a small tangent because you mentioned open source,

I have to uh go to Perplexity

here, who you have been a a fan of a long time. >> I love it. Yeah.

>> And thank you for

releasing open source Neatron

3 Super, which you can also use inside Perplexity. look stuff up.

>> Uh which is uh 120 billion parameter open weight uh model.

>> Uh what's your vision with open source?

So you mentioned China

with with Deep Seek with Minia

with all these companies

really pushing forward the open- source

uh AI movement and Nvidia is really leading the way

in um close to state-of-the-art open source LMS.

What's your vision there?

if we're going to be a great AI computing company, we have to understand

how AI models are evolving.

>> One of the things that I love about Neotron

3 is it's it's not a just a pure transformer model.

It's transformer and SSM.

And uh we were early in

uh developing the the

uh conditional GANs which that progressive GANs which led

step by step to diffusion.

And so um the fact that we're doing basic research

in model architecture and in different domains

gives us visibility into

you know what kind of computing systems would do a good job for future

models and so it is part of our extreme codeesign strategy.

Second, um I think we we

right rightfully recognize that

on the one hand we want worldclass

models as products and they should be proprietary.

On the other hand,

we also want AI

to diffuse into every industry and every country,

every researcher, every And

if everything is proprietary,

it's hard to do research and it's hard to innovate

on top of around with.

And so open source is fundamentally

necessary for many industries

to join the AI

Nvidia has the scale

and we have the motives

to not only skills,

scale and motivation to

build and continue to build

these AI models for as long as we shall live.

And so therefore, we ought to do that.

We can open up, we can activate

every industry, every researcher,

you know, every country to be able to join the AI revolution.

There's a third reason which is

for that to recognizing

that AI is not just language.

These AIs will likely use

uh tools and models and sub aents

that were trained on other modalities of information.

Maybe it's biology or chemistry

or um you know

laws of physics or

you know fluids and thermodynamics

and not all of it is in language structure.

And so somebody has to go make sure

that weather prediction, biology,

AI, AI for biology,

physical AI, all of that stuff

stays can be pushed to the limits and pushed to the frontier.

We don't build cars, but we want to make sure every car company has

access to great models.

We don't we don't discover drugs, but I want to make sure that Lily

has the world's best

biology AI systems so that they can go use it for discovering drugs.

And so these three fundamental

reasons both in in recognizing that AI is not just the language

that AI is really broad

that we want to engage

everybody into the world of AI

and then also codees of AI.

>> Well, I have to say once again, thank you

uh for open sourcing

really truly open sourcing uh Neatron 3.

And >> yeah, I appreciate you were saying that we open source the models, we

open source the weights, we open source the data,

we open source how we created it.

>> Yeah, it's pretty amazing.

It's really It's really incredible.

You're originally from Taiwan

and have a close relationship with TSMC.

So I have to ask

uh TSMC I think

uh also is a legendary

company in terms of the engineering teams in terms of the incredible engineering work that they do.

uh what uh what do you understand about TSMC

culture and their approach that explains

how they're able to achieve this

singular unmatched success in

uh everything they're doing with semiconductors?

You know, first of all,

the deepest misunderstanding about TSMC

is that that um

their technology is all they have.

that somehow they they have a

really great transistor and if somebody shows up another transistor

game over >> it's the technology

and of course you know I I don't mean just the trans

transistor the metalization systems the packaging the 3D packaging

the silicon photonics the you know all of the technology that they have that

technology is really what makes the company special

their technology makes the company special

but their ability to

orchestrate the the demands

the the dynamic demands

of hundreds of companies in the world as they're moving up, shifting out,

you know, increasing, decreasing,

push pushing out, pulling in,

um changing from customer to customer,

uh wafer starting, wafer

stopping, uh emergency wafer starts, you know,

all of this dynamics

of the world's complexity

as the world is shapeshifting

all the time and somehow

they're running a factory

with high throughput, high yields,

really great costs, excellent customer service.

They they take their work ser they take their promises seriously.

when your wafer because they know that you're help they're helping you run your

company when the wafers

when the wafers were promised to show up the wafers show up

you know so that you could run your company

appropriately and so their system their manufacturing

system is completely miraculous

I would say then the second thing is their culture

this culture is uh

uh technology focused on one hand advancing technology

simultaneously customer serviceoriented on the other hand

a lot of C companies are very customer serviceoriented,