In Good Company with Nicolai Tangen - Jayshree Ullal - Arista Networks की CEO (Hindi version)

Episode Date: January 30, 2026

इस एपिसोड में, Nicolai Tangen की बातचीत अरिस्टा नेटवर्क्स की मशहूर CEO जयश्री उल्लाल से होती है। जयश्री समझाती हैं कि कैसे अरिस्टा आज के AI सिस्टम के लिए बड़े-बड़े नेटवर्क चलाता है, क्यों AI का ट्रैफिक पहले के सभी ट्रैफिक से बिल्कुल अलग है, और क्यों अब सबसे बड़ी रुकावट हार्डवेयर नहीं बल्कि बिजली की कमी है। वे अरिस्टा की कहानी बताती हैं—कैसे एक छोटी इंजीनियरिंग टीम जिसकी कमाई शून्य थी, दुनियाभर में लीडर बन गई, वो कल्चर जिसने कंपनी को सफल बनाया, और इलेक्ट्रिकल इंजीनियर से CEO बनने तक का उनका अपना सफर। यह एक खुली और गहरी बातचीत है जो लीडरशिप, नई सोच और AI की दुनिया में नेटवर्किंग के भविष्य के बारे में है।——————Jayshree Ullal - CEO of Arista Networks In this episode, Nicolai Tangen sits down with Jayshree Ullal, the influential CEO of Arista Networks. Jayshree explains how Arista powers the demanding networks behind today’s AI systems, why AI traffic is fundamentally different from anything that came before, and why power—not hardware—is now the biggest constraint. She reflects on Arista’s evolution from a small engineering team with zero revenue to a global leader, the culture that shaped its success, and her own path from electrical engineer to CEO. A candid and insightful conversation about leadership, innovation, and the future of networking in an AI-driven world. Hosted on Acast. See acast.com/privacy for more information.

Hi everyone, I'm Nikolai Tangan, Norwegian Soren Wealth Fund's CEO. And today I'm here a this person who is AI for very important. Jai Shri Ullal, who Arista Networks' keyo is Jayhashri

has a lot of this year's more than any better than any can't. Jayhashri, you look at you. Here here here back a lot of a lot

like, Nikolai. I'm a little sardy so if I'm naq so I'm sorry. No, no. Now, we're going to

start from the first start. I'm going to this podcast. I'm not going to be a lot of so it's not. So, let's

start from the start. So, let's see, Arista, what's the question. So,

Arista, the most demand-wale mission-critical network is making high-performance, high-scale,

low latency, high reliability, high-availability, high-avail-o-o-o-o-o-o-o-tomation, high-telemetry-we-we-old-old- which adhonic data centers and AI needs to pull up you're at a lot of level

on a lot of level on a lot of level of a high demand has. And, especially for in the first one decade, Arista,

this field in pioneer and leader, both of the group of upher and I've seen, if you

you think, if you're that's type chips engine then, then, networking is their

way way is, it's right way I'm, I'm the chips both things

drive Khrase AI's in the world, where you processing is GPUs,

or you know any of the name. But when you do you're going to you have to you need to

do you need to do you? And this the way, if a car speed is 100 m'clock but it

can't be 30 mil per hour is because infrastructure that's much to

do you need to help so you you're you're you're very much

use of a We're all accelerator, users, mission critical workload information highway, which name you give you. So, let me, now I'm in my office in my office in chat GPT on a prompt downed, I'm how to use your products how is this service how does it? Yeah, so we're, we're the other part of the picture he come are.

Like my parents always said, you're what do you? Because they don't see what we actually what are. And when you're Arista's prompt typeed, So, at that point we're basically everything to connect to everything's connected to this podcast on the conversation is without possible

not possible. So high-intensity traffic pattern, chahe they're a user, device, workload, machines, servers, storage, or

now, now, the higher-ealer application to be, we connect and it's switch and routing platform of a set in just a

leaf spine architecture where multiple leaves a aggregate spine connect with and today are these may AI spine be can be able to

this world's big high-scale platform. But, they need to power to need power to support and

chips and power to power to need to and here here here here is Aristas's

especially its extensible operating system EOS is where we where we platform and

networking connectivity to the data to power to power We're structured data, unstructured data, flow data, and these all of their

specific software, to work in work. And AI's networking, the first systems were, how are different? If we're a little bit more 20 years ago, networking to look at, the internet's start, research community, had. That time, TCIP-Jas protocols and algorithms research projects, especially ARP&IT.

This foundation on today's modern internet is the foundation, where research which has practical networking systems role used. There were many protocols like XNS, Apple Talk, OSI, and IP. But in 2000, TCP IP, the world in the world

in the internet communication common language been being. 2,015 to 20, up to, after, just TCP IP is not enough. Because networking

need more complex were complex. This door, people were the world the world's big cloud platforms began like, AWS, Azure,

GCP, Google, and many others. They had so many big level, unprecedented scale, that they were that they could where they're

to get so they're so they're multi-tenant been to be because they're just a company

not and he's many people but he's many people have been able to make sure that has been done and

it's going to depend on so they they're a mission critical assets can't so he

e-commerce application or database or other. Now, as AI in the

last year has been and this new twist is a because cloud traffic is AI traffic is a lot of different. AI traffic, and we all have

chat GPT and other things start with the same complexity and search is a lot of times more than a lot more than ever seen. So, before, ever the same thing to see, to this for this traffic pattern of more, more than a less latency,

and high performance to be. So, on the AIME's before we were doing, it's we're front-end network which standard CPU, storage, general purpose query, indexing, searching, this type of things jrothed. AI's coming,

we're just a thousand or then, if a lothons, if you're connecting, and as I said, as GPU traffic for a lot of traffic need, like, as it's traffic's quality,

traffic's intensity, traffic's durability. peak burst in the big burst in the matter of very intensive is the we're often it's often

back-end network or scale-up or scale-out network we're going to we're not a lot of parameters and

metrics were to until we're I'll say that's the lot of things were like an infini-band

network was a fiscite network, a CSI network and CSI network and Arista AI for back-end

in the internet and IP and standard-based capabilities to bring in the business in the world in the network's

in the now the now the the I'm I'm I'm

the most big power is because in GPU and and and optics and all and all things

power consumption not a lot first we we were about we were talking about we're at we're at

we're talking about we're see the same right? It's the football field's

big that you're and you are you and you are the power so that

our AI provider or cloud provider is this level power not if

you're if you are you're you're you to look you're

you're you're three to five years can't be the place. The Gaguer and probably

still the big problem. But one you know, if you're you're not you're the same way to

you're the same way, we're doing you need to call to call have to be but you now are

a high infrastructure is a lot money is how you see it? Do you this is a

way? Is it a way? Yeah, this is the badal over the a very teasy's

so networking the working the current cycle to 1GB,

then 1 gigabit then 10 gigabit and 100 gigabit to reach to get back

every 5 years. But now 100, 400, 1, 2, and 3.

2 terabit to migration every bar in 12 months in

here are that. I mean the bad the first in

the first 18 to 18 to 18 to in are in.

It's meant is that we we've we've got we've got we've got this year we're

this year we're first of the year we're going to think of the speed scale is a amount of

because you have you have all you have all the throughput and unumannet latency for all all the same you're all right

you're in I was in I was I was you, I was you know, you know what you're doing it can't

how can't how much it's a lot internet boom very, very sthir way was a time

was a peak in 1999 2000 in 2009 2000 which was that was and then you and then

Y2K but then it's the internet to develop to develop to be 15 years.

I was I was in 1903 before acquisition from when the company a billion dollar's

came and when I did it 20% in 20% billion dollar company was

where Catalyst Switch, which I think that still the best product line, at least a third-yogad do they were but the part was that it

was a little, and it was a peak was, a week for two-sale it was a way of a lot is a lot of a lot of

the reason why you were being the first-and-a-and-a- thing, they were up sikud-as-sard in five-sard-sall-same in-same-a-same

and over-a-old-old- -one-old-old-old-old-old-old. It's like-houser-gggguna-old-rerereys-old. And the third part, I'm the most important, because people are this to look at it, is the problem

are we're in I'm in Bubble in Merex. I'm a lot of we're in a lot of a bubble in, we're in a long bubble in. This CLEX

just companies who came, they're funding made and then banned-ho-gained. This Microsoft, Google, META and Oracle like a similar

companies who are they're taking them to make them. It's also, new clouds be, like Open AI and Anthropics.

which ishacr to do you. So, I'll say, it's a bubble than a great-a-trend like it's a

time to here's going to be a lot of time. We're going to we have overbuilding so we

can't how we can't be that's not this is up-uped not so up power and not

this this is a time will be time it's a our aim it's over it's

a ability to where's the births. So, I'm not the nine-09-199. When all the people'd make

companies, and they'd say, we'd say, they'd be able to be able to do. This birth situation is all too. Here,

this is the first need to be the need to be the same big problem. But again, it's not, that I'm not I'll take not

and I'm not say I'm that it's a bubble not, but it's saying it's very easy and a lot of a

and a lot of a matter of But it's I know, it's a steady state is a way that

we're all have been there's a but I'm not think it it's a year that's a year

and it's this thing and these companies are doing they're doing that they're they're

they're doing they're the the same the same the same big answer. So,

you're the people, so you're the client hyper-scaler are there, you know,

cloud companies, Microsoft like companies, yeah, so customer customer base, how concentrated

are? Well, they have always concentrated are very as an example Microsoft

and Metta 2000 in 2014 public being public after the revenue

every each per percent and I not I'm this

concentration for not I'm I'm happy and I'm

so they have the revenue of 10% been but it

but the the denominator big is the We're very speciality cloud providers enterprise campus in diversify

but if I'm not this not tell you, they're not really and partnership never not really much much so it's a mousathe. When AI capability to make the part

is the best impressed? Which country? No, no, the United States, and we lead, we're doing. Doosieres in

other countries are going to but in scale in the same deal in the world's in the world. And the companies we're working in those, most part are US in. Is there any other other country that you're doing? I'm like, I'm like it's

some of the Nordic nations in the Middle East in the UK may be some of it. It's a little bit of Korea and India in. So, it's no doubt that it's many other international locations

to the same to reach to the when you see the investment where are so you're

you're going to you're going to you're going to the world before the I'm going to they're not they're not

more than they're more than they're more sponsored not so they're they're all they're all technology and

communication to how much about how I'm going to I'm in middle east and I'm there's a

progress of Raffaar's very impressed where they have been very high investment and investments and

are there's so I'm it's a kind of a bit of a kind of look at a impressed

being about you're being, you know, you have you have a AI data center networking for market

market's a part of the you're very to get to how you? How are to we're just

just that we just need to it's just it's a lot we're more absolutely to make sure to have been

is that if you're high-speed market to look at least we're sure we're about you're seeing the same speed we're maybe you're

that you're about you're looking at least. This is this is a company 2,08 in the company was made

and it's not a lot of we've done it we've done it but putke to work and I think this company

culture to optimize to optimize to endy Bectualsheim and Ken Duda had to be able to they've got our software, our hardware, infrastructure for the same name for it's better

to make it. So, our market share actually is the part from the cost of our innovation to appreciate. We're made by engineers by engineers for being made. We're quality architecture and we're very time and

support them support because we're a mission critical mahaled in a center where we're downed then, then we all know it's all over there, if we're down to, we're the world's

the biggest big a and I'm not a lot of this in a big thing in I'm a methodical

way he's a and we're a product not used that's a thing that's before he in the bazaar in the

launch-gared you. Now, you've spent 15 years to make, their data center business

to bring to so I just just know, that's just how to work to work to work.

It's just just a data center business's I'm much, I'm a same time I'm a

When I joined CISCO joined then the company revenue was about 650 million and market cap 6 billion was.

This is it not a year-up not, but a youth company was a new-a-company. I got a moment and the same

company's many many the same during I made, I'm making, and develop to develop

to give down to give down. First, this Catlist was 1,000, 200, then 5,000, 6,000,

500, 4,000, 3,000. So, this switching brand when I was the first It's the first time it was zero. And I'm today,

I remember that the first year it was 40 million, then the second year, and then the third year is one billion had been, and this sort of things worldwide web and

internet can't even if your product, why, why not have, I thought I a rocket ship a hissue on the

great, it was a great company and it was a but I was I thought, I'd like, I'd be a big company

in any big company I didn't want to be in I'd want to try to but I'm a single-in' but they're 15 years

really good yeah. What's you know? What's you know, what you think?

I'm not here here here's the thing, you know, customers on customers focus.

CISco in customer focus culture is. I'm that it's on innovation,

engineering and quality on the quality on but I did it. But I've said to

night two when I would have been very, then I'd be the phone would have, you know, if I'm very important, but that's

important, that's important, that's the way that's the culture is very important. Now, you're

maybe Cisco's the CEO can't be the you, so you want to be the ? I don't know,

I'm not I'm I'm a Sisco for Sisco for I was a CEO or not,

because I Siscoe was an accidental executive was. I never planned not I didn't have

I'm the executive I'm going I'm going. I'm I'm going to I'm, but I'm,

I'm like, do you know, I'm not many people are not there, I'm not, it's not, that you're

you're going to think, hey, I'm Sisco, I'm a superstar going to be going to

go to be a you know, you're you're not a time I'm five-sale to me, but I'm a lot of

this is I'm going to I'm a product was a, I'm engineers with work with customers for

products to make a matter of I was was a lot I'm not thought I'm like I'm

more big capitalities or Cisco in CEO being a issue is so

when I actually I thought I thought I I'm a different I'm a lot I'm

like clean take or energy or solar or battery and I really I actually in Silicon

Valley back and I'm back which I I was I mean, an entrepreneur and I'm

Arista in when I'm in when it's there's a real real estate and there's the first thing that I'm

that's the thing that I was that I'm that I'm doing that's I'm a money to let's a

startup is my revenue zero 30 people in a office in a so I'm you like you're

so I'm how I'm how I'm I'm I'm I'm not I'm I'm

it's Andy and I, this is the time. We're one of the other we're doing. So this Andy Bechtolheim is? Andy Bechtelheim,

who company's German co-founder when I was Cisco in, we had been their company Granite to and that we had done

we've done with other way. He's the other than he and I was a MECrocy and I did you, and I'm a

person who's one and 18707 to know that's more than more than some more than so much than he looked at

When they saw, I'm taking and I didn't know what I'm going to be the I just my mind safe, I'm just

my mind there's a start-up I said, oh, and my first reaction was, oh,

no, Andy, I'm still networking and switching and I'm but as the

founders, what were trying to and the engineer what were making, I'm

I'm interested because it was networking and switching but but a twist with a

internet and internet and catlist switching just I knowed this is a

different thing. It's what was that's you know that's a lot of

this thing. It's software and architecture very unique that's

many people do you have they have five use case for five operating

systems and every each one and many many hundred things

that's only only one only binary image. There's

there's there's some state-driven it automatically recover and the agent back comes out of and it was a

Christmas tree light where one light after the light's end up and all the lights we'd be able to be able to do you. It's a system that's just the LED bedal did it, and the whole system and people but,

but I'm I'm a little bit of the year was a year 2008 was that time financial collapse

was our our customers from one of the and brothers were what you know? I know

So, Arista in my entrepreneurial career of the short not really now is

that now is that we're not that we're we're not we're all over-chard-chard- -chard-and-bent financial collapse one of them

one of the other competitors, Cisco, to him on case to us other part was the COVID and the whole supply chain's also a lot of so we're going to we've got to come and here to be a lot

good. What do you need? What do you need? You're a start-up and think we're going to have? So, what do you think?

I'm many colleagues you'll tell you that I never thought that I'm justicco to takeer to be a Because in the first we were very unique use case that we were focused that Cisco was not not even being involved in it, or to do not

give them. Like high-frequency trading and low latency. These were such use cases which in Cisco's not only care of. And our first $100 million revenue, the most of here from. Our second use case was Cloud, and Cisco, it's MSDC,

which multimassively scaled data centers, he said but he didn't even didn't even not even that's the customers

what are the customers, what are they do they're in the actual way, they were they were

not to CIS COESCO or not he Arista because they they had they they had

their system made, they had they were they're they're very disatisfaction

to dissatisfaction so. So, I thought, we thought we we're a we're a

kind of So, big corporate corporation, especially Cisco, that same, that time, are they? Behtarine customer service example is.

As a many examples and many can't you. But let's see, I'm the same the same question. If you

if you're if you're you call then you think, what do you? And, let's know, and you

have a network problem and you'll call to you can't you can't you can't expectation

that you'll be able to and you'll get to get the same, we're in often at a question,

you're going to you're going to you're going to say, it's not a there's not you're not, but we're

we're going to ask what the problem is, our average resolution time very high, it's a way,

it's a emergency in an emergency in a marriage to a lot, you're you're not questions,

you're you're still you're symptoms, you're we're So our average resolution time 25 minutes. This is an unseuna record

because our team experts are brought and quality and support and along along. When you're high quality products make, if any other

configuration or support problem a, then we turn to action and we're there's just we're outsource not we're outsource

not. Our follow the sun model is, worldwide coverage every time. We're just response we're we're really response

and it's solve and average time quite come and it's a high-quality experience is a very high-quality experience is. How do you

how do you make sure you? It's a little personal time like it. I'm a calendar in a time

I just just I just think on and no meetings not because everyone who everyone

I want to me meeting to be I'm always I'm updated to be in a good

example for AI 2012 and 2013 in I made I've got I'm update because

I'm I'm not paying I'm not doing it and when November 2012 2012

was the CHAPT moment I thought that it tech industry for a

big milestone was a mind I had that I had I'm in January 2013

in optical conference OFC to talk to do there I optics long haul,

short hall, DWDM, co-packaged optics, DSP, whatever, about I'd talk to be back but I'm

this to be the conference use, AI, to discuss to a driver's to discuss to do.

And, this is to make, I got to because you are you talking from

PhD's audience to maybe maybe I'm more than me's

more know, and this the way I'm very very good always seem to be

a lot of a lot of people, you know, in the industry in their saithy

and Arista and there. Our organization in many top technical experts are you. You can't

how much can't do you? It's not . I can't be a way I can't

be a lot of things can't be a lot but I'm some good accuracy one to three years

kind of kind of other a good situation and then what are what are the next year?

I'm I think that we're the way that we're going to that's going to be a lot of we're going to

a AI's a frame version are we're about big clusters clusters of crores, and carbom parameters and

tokens of we, we see that we all these all terra-flops, terabts, or terabits capacity in

how it's see it's but I'm in the three years this is very important

I'm not not quite, I'm not I'm going to move to be a more distributed problem will be

we all we're we're all same same fit not and just Menframe client server

and after after after you know distributed PC and client and phone on the way the

way I think I think AI also I can't many different shapes and sizes

in it and you know this in this way this one is this and it's what implications

are? It's a significant. Because miniaturization is the big and the power system

to make a way more much difficult is it. It's the matter we need to reigning and

inference in translate and we have to and in many use cases in some resources in a

more resources, because if if you if you if you're if you're just bandwidth, power and

the biggest training workloads to charge. But if the capability in reasoning in law, using, distributed

way, general purpose compute and storage to lay to get to , the AI, hard desktop to reach

to reach up. Just big and powerful system to be limited not. Do you? Do you?

Do you think that? Technology's, the world, America and China and between two

parts of going to be able to be able to Today we're, ah, we're just the way we're

a political pehh wish, I'm not innovation and technology a small better than

field but I think we're all different nations are shasit and so I'm

I'm worried to be a good step up to we're we're we're always innovation

and we're always best talent with partnership have you have been you know,

I'm a I'm like an immigrant example who here came here came and here didn't have been in China in, whether it's in China in,

or London in or we need to get access and we need and we're bettering with the same need to work

now. Now, culture, how you how you describe can't? Arystah's a good

work, because if a public company on the only thing on a thing is easy to focus to be

easy to be there's a business model has a business model and that we have focused to focus we're going to

our own work our hardware and a chip member with a lot of a lot of quality problem. We've done with them and they've worked with us

and I had just software in a little patch to make a option but I made a full question and I made a pretty question and I've got to look you're not looking at

but it can't be able to can't and we're on our lot of your place

you're in this it's not we're back up but we're going to do you know

so people who we're doing that's the bad that all the other than all all the

good other so I'm so I'm a company to work for a work for

focus to a culture to our customers for the that is that I want to be that I'm

you can't you? You can't make sure do you make that this attitude in the attitude in the

organization share to this is it's a easy to you're going to get to be but you

but the thing is that's the management team to know, we're we're

we're we're the management team and we're sometimes people people

lose other other luckes or financial success because

but I think you you're you have you have the management team and

leaders are and now we're we're we're we're we're

are you are bringing and you are you make a lot more mustbue so it all the organization

and I'll say that we have been we're not we're we're we're still professional

manager not but they coach and team's players and every and every role

actively and give my president and CTO Ken Duda who is the company

founder also they they're they still they're so he our carmcharies to them to tell them

we're going to say, we're going to lead are we're we're going to say, we're not going to

do you? How do you? She'd I'm not not really see't. As I said, my seekne

sources are many things. When I I'm a customer to be they're much teach out of because they

they're a different things. This is a time. I'm a other engineers to be

many time to know, he's what they're and they're what's the things and they're

they're always me something or something something or you know to find out you know

or Jemini or Chatt GPT and you're very efficientity with a software developer can't be who complex

topics to understand easy I just I just in my just in my

small but it a good source of learning and and learn

my my most best the I'm much I'm I'm

but I'm but that's my a very much a different way. Jashri,

when you do you see what do you see? Yes. So, it's you know, you're very

safe is it really safe is that they're a good, and there's a element and there's about their

other to ask them to ask them. But I see, I'll see, ability, character,

moral values, culture, and what kind of person is, is, you're a good person

because you know, I want to I want to you know, that this hire not just not the result of

can't be but long as long as much to be and we're going to our culture

and we're in the kind of and then we're back on back on we're

so we're we're not long term about we think and so we're

and so we're who's who's such a kind of who people are

if you if you you're a lot of apotism but I'm my mind that it's a

good people because if you if you're the employee know, then chances there's that there's

that's a reality of that's if we're having to I'm always I'm always ready to be

you're always people. How many percent people are Indian background are? I'll

say, because our India in India there's a big facility is so it is this year

it's about 15 to about how much we're about we're in the world, you're in New Delhi, you'd be able to? So, when you'd

what do you'd like see? Yeah, my janu london in and perverish in Delhi and I've lived in United States my father was

he's a teacher, so they were they'd think that they'd like I'd do you, they'd first five IIT shrewed,

as you know they're very much much and a good reputation of, so I'm, I think that middle-class

in the institution education is a big big a lot of doing, you know, we know, we're very good to

just a good to do, school in school and go to my mother-pita did that they made me to

they can't make a so I never never seen. I never seen that I'd I'd be a one of

I'd take how school how I'm how I'm how to how much how much to come

to come to that time there was there was there was But I knew that my mother-pita was more than much more

ummied. Shai'at, I was there I'm not too much ummied. When there's so it's a lot of what's the effect.

Sometimes it's it's a lot of it's a much more it's a lot of it's a very interesting that when my father or mommy parent-teacher

association in my teachers to meeters to they always say, Jeshry in there's a potential but he

but he's a full use not using he's not using this is a and he they were

and they were more more of the and more umed but it's a special seek made

my very a very a very a very group of in India like students'

ranking had. He always first had and I never first not I never came

and I did that he know in these exams in a chance to for a good-in-a-up in the day,

I'd like eight-gontes part-ganted and I'd in a day- three-four-gonts he'd be a lot of a lot of

if you're if you're saying, I got a that was a good-of-a- but I'm a top-10

percent in I was in my parents to make sure I'm trying to I'm playing and I'm right-you-

look-sacted but that that time this day in eight-guntes part-and-a-mach-a-d-mah a way-a-a-hanna a lot of you know,

that's the time that was the time that was the time, not a lot of then you know, I'm going to yeah, I'm

there, elimination of I'm going to physics, I'm chemistry and I'm math, I'm not

, I'm , I'm , I'm not good, I'm not , and I'm thought, what's the thing is,

that's, that's, you can't give me, and computer science still, a quite a very rare

field, in my time In the way if that if that were the offer if I'd would have been I'd have my computer science

classes in math department to learn fortron C++ and assembly for I'd some classes

but electrical engineering was actually hardcore engineering and now it's again it's again

this modern world. I'd rather silicon and chip hardware design in so people

so I'm I'm I'm I'm very my father's who had IIT shrewed did the

who had did the science and real world of work that's that's the engineering

was that that was that was that's a much that's people who you're you're going to

you're you know, you know, I'm I think you you're you're not you're

so it's you're not you're you're doing you're you're you're doing it, you're

you're you're you're As a CEO to the same the same thing is. And I'm a very money

my mom's the same you're very you're very you're going to I'm going to I'm doing that's a so I'm

never a lot of many of the money. It's always it's very it's very that I my work

to my health and family to balance but I but I'm such I'm I need I need to

maybe I maybe I'm maybe I'm maybe you what are? What are you? What are you? Have you?

I was, I'd I'd probably I'd but I'd get into my hikes are I'm a lot of my dog is a lot of

music's really I'm maybe I'm not a plan but it's a but it's a lot but it's a shower in

the voice is kind of there's a other can't be a lot and now now today now today is

it's more be a more than than I'm I'm really I love I love I'm very

very much like to You know? What do you? I'm a lot of biographies I've got to. HALM I've got not not seen.

But when I'm, I'm the most long-distance flights in during he puttie. And, or I do some specification

print and publish and some technical document or I'm, or then I'm a little full-nauvel like I'm,

like John Gresham like, which in it's a big, it's a mind-in-mere that's,

that I'm to say, I'm more more than more than more, I'm a lot I love

I love I'm the first the part is the first thing is my family I'm the

I'm the my wife and my brother and they're much much they're because

that's not that not that they have not really I'm my alma mater

Santa Clara University from honorary doctorate degree but I made. But my

my beti did such meh made PhD and one did a veterinary science

in doctorate did. I'm I'm in their values for gerv and their compassion for the

and this thing that they have been a choice of you know, youba people who in your women who

do you know, I'm often I'm often that's the thing, because all the people are

many people that's very much much to get a lot of much, I'm I'm not a way

to make sure on, I'm every person has a talent has a which is

we're things in some things in some things in some things in but very bad. But if you

can't you can't which you're good, you're able to make a better and you're and you're

working and you're working and I think that everyone better being kind of it's always just pay for

not to be mappas. It's just the most manpacion should you you're you're suhhh

you. Jashri, so it like that that's like that you're all right you're all right

you're all used to puttation to use the world is all the world to do you know, you're doing to thank you

and for the other to you're very good good good news thank you Nikolai I'm

so much thank you thank you