Hello everyone, my name is Raj and I'm

recording this video to share some

honest thoughts which I have been having

about how things are evolving in the AI

space and uh

some

recent conversations which I've had with

my friends and I just want to express

what I feel. So in my batch of IIT

Madras

around 80% of the friends who studied

with me

uh are in the United States or Europe

they went to study for their masters or

PhD then they got a job in Microsoft or

Apple or Amazon, New York Times etc.

Many of them have now risen to the ranks

of

um senior AI researcher or AI lead at

these companies.

Some of them went into consulting and

then went into private equity and VC who

have worked so much that they are now

experiencing feelings of burnout. But I

can confidently say that about 70 to 80%

of these people are not in India right

now.

If you contrast this with my journey, I

came back uh here in 2022, 5 years after

my PhD at MIT. And at that time, I used

to feel insecure because I had left

everything.

Um I was getting a good job at Amazon

and I could have even joined the Julia

lab as a machine learning engineer. I

left everything and I came back to India

to start Vijuara.

And I was very insecure at the time

because my friends were all rising up

the ranks. They were in these big

companies.

Uh but and I thought that ultimately

okay they have their own life. I have my

own life. Everyone is following their

own paths. But these days you know um I

have some very strong feelings because

AI agents and the rise of claude the

rise of the rise of claude code

especially opus when it came have

completely changed things. I I'll give

you an example.

Imagine that you are working at Apple or

you are working at Amazon

and you are a senior engineer there who

has put 5 years of effort and you go in

there at 9:00 a.m. and you come at 6

p.m. The fact is that most of your work

you use coding agents. So you probably

have 2 to three hours of work which has

been assigned to you by your company.

Then you have you get pulled into

meetings, managerial meetings or you uh

designate your task to your team members

and you are completely cut out or you're

completely detached from what's

happening in the world around you.

Whereas the reality is that the world is

moving extremely fast as compared to the

silos in which you are operating. For

example, if you are a manager, technical

manager or a lead scientist at Amazon or

AWS, why would you care about learning

how the Deep Seek harness works from

scratch?

Why would you care about building your

own JV model fully from scratch? Why

would you care about hosting Kim K3

yourself locally and testing out its

implications? But the world is moving so

fast that I now feel that these big

companies have literally nothing left to

offer engineers.

They have nothing to offer. Amazon, AWS,

Netflix, Meta, all these big companies

create silos of jobs for a person and a

very highly qualified person fills that.

For example, you might be working to

optimize ads revenue for Apple app store

and you do 3 years of your life

optimizing for that Apple ads revenue.

Why why is it necessary to do that in

the in this age where you could

literally start your own company with

the help of agents and build anything

which you want. In the morning you can

study about the deepse harness. In the

afternoon you can build your own jev

model. By the evening you can set up

your own agentic company pursuing your

mental health startup. All of this is

fully possible. Imagine that if you see

some if you are always passionate about

mental health right and if you are stuck

at a company right now and you don't

know what to do because you are in a

grind $200,000 per year are coming into

your account. But I'm telling you right

now that if you leave aside these

$200,000 a year and you start your own

company, there is so much freedom now,

you can literally do anything you want.

Coding is not a restriction right now.

Design is not a restriction right now.

Marketing is not a restriction. Sales

outbound is not a restriction. If you

learn how to use agents correctly, if

you learn how to use agents

appropriately, why not start your own

company at this moment? Why not start

your own mental health company with your

wife or with your friend and just go for

it? If you are in the US right now and

if you have spent six to seven years

working at a big tech firm, you probably

already have savings in your account,

right? So just quit the company and

start start your own thing. It's the

best time in the world right now to

start your own company. A company which

can be fully run by agents or have some

humans in the loop and just

go ahead and build a minimum viable

product and try to do your first sales.

Now the time to build the minimum viable

product is not as high as it used to be.

Um

let me let me give you another example

of one of my friends. Let's say you are

working in Microsoft. Okay. and uh you

have been tasked with implementing or

improving Microsoft search engine but

you have seen that the company never

uses generative AI that much actually

that's one thing even for you to deploy

a new open-source model you have to go

through the

um infrastructure team you have to get

their permissions for 3 days so you are

stuck in a role which is not modern

because they are not using generative AI

you are stuck in a role in which you

don't know about modern things like

let's say Hermise or a pie. Have you

heard do you know how harnesses are

built from scratch? No. And secondly to

even do something new you have to get

approval of the infrastructure team. All

of this points to the fact that you

might be stuck in something which is far

behind. One thing I'm pretty sure right

now is Indian and Chinese small

companies which are moving very very

very fast have a huge advantage compared

to these big siloed open companies which

might move slow. Consider Vijara for

example right in the last 1 month I have

done so many things which I could not

have done at a big company. I have uh I

have designed the whole kernel

engineering workshop fully from scratch.

We have partnered with an enterprise uh

on a multi-million dollar deal with them

and we are solving uh AI problems for

them. I have written the build deepseek

from scratch book with Manning and three

other books are in the pipeline with

Manning publications.

Um I'm co-hosting webinars with some of

the best researchers in the world. We

have three products built at Vijuara.

Vijara books, MW and Vijara pods.

It is just incredible. Then next week

I'm going to IIT Madras for a podcast

with Professor B. Raindran. All of these

things would not have been possible if

I'm stuck at a company right now. And I

get it right. If a new model drops or if

a new technique drops, what's the

incentive for a person who is working on

uh the meta ads feature, let's say,

what's the incentive for them to learn

about it? There's not. As long as the

monthly check keeps coming to your

account, you are probably fine.

So I feel that this is a time where

it's very very very high time that we

really reconsider what we are doing

because once you quit and if you see the

world which is opening around you

everything is open right now every

single thing is open I'm sure whatever

you are doing in your company right now

you can start your own startup based on

exactly that same thing and even get

your own customers and just build your

own company I mean what's the risk here

you can launch in 2 weeks You can launch

get your first 10 customers. If it

doesn't work out, you iterate. If it

still doesn't work out, you again

continue iterating. Already you have

money stored in your accounts. So just

go for it. What's the point of working

for 2 hours in Apple using clawed code

and then just whiling away your time for

the rest of the day?

I work 10 to 12 hours every day and I'm

learning new things. I'm doing new

things, developing new things which are

of my choosing. I'm fully independent as

to what I do every single day. No one

dictates what what I can do, what I

should do. My time is my time is how I

decide I need to allocate it.

And I'm not saying this lightly because

I did not make this video earlier, but

now I'm I see many of my friends

actually leaving their companies to

start their own companies.

All of my friends who are in the US now

want to start their own companies or

start their own startups

but many keep saying this and never

never take that initial step. I feel

that now is the time where you can just

leave your company and get started

because apart from a stable paycheck you

should ask yourself what is it exactly

that my company is offering me. If you

see these uh hackathons or fairs of what

China is building right now, you will be

truly amazed at the pace at which the

world is going. Small Chinese labs are

so ahead. So just keep an open mind and

don't rely on that monthly paycheck.

That is what my thinking is right now.

Start your own company who who knows

where it goes. But one thing I can tell

you is that earlier if you thought that

oh I am a nerdy type of a person. I

don't know how to do sales or I I I

cannot make a sale. I don't know how to

do marketing or you or if you're on the

other kind where you think that I've

always been a sales and marketing type

of a person but I don't have the

technical chops. I'm telling you now you

can make use of opus 5.5 sonnet 5 codeex

who can do those things for you. Sonnet

can be your technical coding counterpart

but opus can be your marketing

specialist.

It just you need to know how to use

these agents properly and once you know

that then sky is the limit for you

just keep an open mind because if a

company decides they don't need you they

might lay you off very easily instead if

you yourself decide that maybe I don't

need my company anymore then you can

take the decision for yourself.

So basically right now I'm uh

vacationing at a resort and uh it's

Monday and Tuesday. So I can take my own

holidays whenever I want. That is the

freedom which you get when you start

your own company. Then you might say

that I could have started or why didn't

you make this video 5 years back because

agents were not there. The arrival of

agents have has completely transformed

uh the economy and I think big companies

literally have nothing to offer now

apart from their name. Earlier they had

something to offer because no one knew

how internals work, no one knew how

infrastructure worked, no one knew how

Apple deployed something to production.

But now with the help of Opus or CL, you

can try to learn many of those things on

your own.

Nothing is a secret anymore. models are

being open sourced and opensource

community is only going to grow

henceforth. So decentralized pockets of

small companies are going to emerge in

the future. More and more decentralized

small one man one person two person

companies are going to emerge and uh

people are just going to have three day

weeks and relax for the remaining 4

days.

that is going to happen in the future. I

think so. But I think this video will is

made for those people who are getting

that $200,000 or $100,000 paycheck every

year and they are just not able to leave

but they know they are stagnating.

>> Whomever I have spoken to, they know

that they are stagnating right now but

they just don't have the they don't have

that push that okay let me go out and

start something of my own. If you are

one such founder or if you are one

person who stuck in this loop just email

me and then I'll just help you get

started

um because this decentralized single

person, two person or five person

companies are going to emerge and I

strongly think I'll say again that big

companies don't have anything to offer

anymore apart from the stable paychecks.

So just email me if you are one such

person stuck into these roles and there

may be a founder inside you

>> who may be waiting to come out. Now

might be the perfect opportunity to do

so. Thanks guys.