Harnoor went from repeatedly failing Microsoft interviews to becoming a Software Engineer at Microsoft but when he entered the Silicon Valley job market again, even Big Tech experience wasn’t enough.
After applying to more than 200 companies, landing 10 interviews, and getting rejected from every one, he discovered how dramatically the tech industry has changed.
In this conversation with Harnoor, we break down his journey from India to college in the US, working at NCR and Microsoft, leaving Big Tech, taking an eight-month career break, and eventually joining an early-stage AI startup.
💡 Harnoor shares: ✅ How he got into Microsoft after being rejected multiple times ✅ Why his compensation nearly doubled after joining Microsoft ✅ What it was like building products used by more than 10 million people ✅ Why he left Microsoft after three years ✅ What he learned during an eight-month career break ✅ The reality of moving to Silicon Valley without a job ✅ How 200 applications resulted in 10 interviews and zero offers ✅ Why startup interviews can be harder and less predictable than Big Tech interviews ✅ Why LeetCode alone is no longer enough to get hired ✅ The difference between working at Microsoft and an early-stage AI startup ✅ Why system design matters more than memorizing programming syntax ✅ How AI coding tools are changing the role of software engineers ✅ Why engineers need to understand scalability, infrastructure, and distributed systems ✅ Why cold-applying through job portals usually doesn’t work ✅ How networking, tech events, LinkedIn, Twitter, and Discord can increase your opportunities ✅ Why learning to code is still worth it—but only learning to code is not ✅ How developers can remain valuable in an AI-first job market
📌 Chapters 0:00 Intro 0:40 From India to Microsoft 2:35 Working at Microsoft 3:19 Why He Left Microsoft 4:48 Taking an Eight-Month Career Break 6:14 Microsoft vs an Early-Stage Startup 7:46 The Brutal Silicon Valley Job Market 9:53 Big Tech Layoffs and the Impact of AI 11:42 Why System Design Matters More Than Coding 12:22 Why He Chose an AI Startup 13:46 AI Agents, Automation, and Security Risks 16:04 Rapid-Fire Career Questions 17:20 The Biggest Mistake New Graduates Make 18:10 What He Would Do If Starting Again 18:49 Is Coding Still Worth Learning in 2026?
👋 about me I’m Maddy, a senior software engineer (prev. at Google), with prior internships at Microsoft, Morgan Stanley, IBM, and Amazon. Sharing my journey here - thanks for watching 🤍
Hanoor, a former Microsoft engineer, discusses his transition to a Developer Relations role at an AI startup. The conversation explores the volatility of the Silicon Valley job market, the shift from syntax-heavy coding to high-level system design due to AI, and the importance of networking and foundational knowledge in the modern engineering landscape.
Most people think getting Microsoft on your resume means you're set for life.
But in today's tech market, even ex-big tech engineers are getting humbled.
It's extremely hard to land a job in Silicon Valley.
I thought, like, Silicon Valley is the land of opportunity.
But it was absolutely opposite.
No one is coding with hands.
They're using cloud code or agentic terminals to code.
The core skill now, which is engineering, scaling big systems,
scaling from 1,000 users to 10,000 users.
So the question is not, should you still learn to code?
The question is, what should you learn now so you don't become replaceable?
Learning to code is still worth it.
But just learning to code is 100% not worth it.
Hi, friends. Welcome back to the channel.
Today, we have a very special guest, Hanoor.
We'll talk about Hanoor's journey from India to doing college in the U.S.,
to working at Microsoft, to joining a startup,
and all the insights and learnings he's gleaned in between,
as well as his thoughts on the current status of the job market, AI, all that fun stuff.
So let's get into it.
But first, I'm sure a lot of people here are curious.
How did you get into Microsoft?
What are some of the resources that you use?
That's an interesting question.
So for me, Microsoft was really hard.
I applied like six times, interviewed six times, and I failed every single time
because I was never good at LeetCode.
And I hated doing LeetCode problems.
I hated sitting down and struggling with LeetCode problems.
But what happened, LeetCode came to our campus,
even though I did not go to a top 100 school,
like other friends who have worked at Microsoft with me.
But I still got the opportunity that Microsoft came to my campus,
but I could not clear the interview.
I did get the interview, but I could not clear it.
So first time, it was on campus when Microsoft was coming for internship.
I got rejected.
Second time, again, internship.
I did interview, got rejected.
Third time, it was for full time, and I got rejected.
But I didn't give up.
So I started working at a company called NCR,
which was right near Georgia Tech and Georgia State University.
So I worked for NCR for two years.
So while working for NCR two years,
Microsoft DM, and I did not want to work there at the time.
I was really happy at NCR,
but Microsoft was creating a new campus in Atlanta.
So I was like, let's give a shot.
Let's see where it goes.
And then it suddenly like, like it,
I got lucky that 2022 was the year where,
when job market was at the peak.
Because the amount of overhiring and hiring that happened in 2022
was at the peak.
And with just LeetCode basic and medium skills,
and that, that's the kind of questions they asked.
Basic, medium, no hard question, no dynamic programming.
I was able to nail the interview and get a job.
And that's how I got it.
Nice. And how was your experience working at Microsoft?
What were your most favorite and least favorite parts?
First of all, when I was actually not even wanting to go there,
first would be the pay,
because my pay actually increased to double.
I was making 80 to 100K a year at NCR,
and it almost doubled the first year living in Atlanta.
And that made me really happy.
Everyone have subconscious out there money,
a big factor why they get into tech industry.
And number two would be,
I think the work was really interesting.
I was coming from a mobile app development expertise.
So I wanted to do mobile app development.
And this time I got to work on an app
that's used by 10 million plus users.
That's called Microsoft Defender.
So that was really fun getting,
getting my mobile app development skills brushed up,
iOS development, Android development.
And that was one of my favorite part.
It seems like you had a wonderful time working at Microsoft.
Why did you end up leaving?
So I left like after three years of working there,
just like you left after four and a half,
but I left a bit faster because we're living in a wave of AI
where like people are happier leaving college,
leaving their job and working in an AI startup
than ever before.
This is a really unconventional path most people are taking.
Like any college campus you visit,
the cloud that people used to get
of working at big five companies like Microsoft, Google,
people used to be like, wow.
But now that wow comes only when you work at like a big AI lab,
like Anthropic Open AI,
or you work on a cool startup.
And that's the wave I wanted to get.
I wanted to be the cool guy
and I wanted to see how it's like being a cool guy.
And I wanted to work on the latest technologies
which aren't brushed up yet,
but will become the future of tomorrow.
So by working for an AI startup,
like the startup I just joined, Hydra TV,
I'm getting to feel the wave.
And I'm actually like enjoying it very much
because I started to hate engineering as well.
Because I think engineering has like started
to like kill my brain a lot because it's a struggle.
When you do engineering, you can basically do nothing.
It's the whole day.
In the morning I wake up,
I think about bugs and problems I'm facing.
In the night and during my shower, during my sleep,
my whole attention, whole day, whole week
is gone towards engineering.
So that's why I wanted to do something less intensive
that can give me a breathing room to focus on my passion.
After working full time at Microsoft,
I took eight months of gap.
I wanted to sharpen my ax before using it.
So I was sharpening my skills by taking a gap of eight months,
quitting full time, traveling to cool countries
like China, Vietnam, Japan, Korea, India, et cetera.
I lived there for a couple months
and I wanted to have breathing room.
And then finally, finally, finally,
work on something I really enjoy.
That is a dev rel role,
which is not as intensive as engineering.
It's so great that you took that break
to be honest as someone who,
the last time I took more than a two week break
was junior year of college, like third year of college.
I think it's something that I am long overdue for,
to be honest.
Yeah, I think there's so much to be said
for just like resetting and like doing,
following your passion.
So you've been working for like four and a half years,
Google and now like almost six years of full time experience,
but no breaks.
Okay.
My current company forces us to take two weeks off
for winter break.
But besides that,
like the last like voluntary time I took off
was third year of college,
which was like seven years ago at this point.
Do you think having the need of room to sharpen the ax
before using it?
I think I should learn from you.
I personally think that is,
I was thinking like,
oh I could like multitask and sharpen while working,
but I feel like sometimes it does get a lot
and sometimes, you know,
everyone including me burns out.
So I think after talking with you
and with a lot of friends,
maybe I'll do that at some point.
Not this year, maybe sometime.
Got it.
Going back to what you said,
you are currently at a startup.
How does that compare with your life at Microsoft
in terms of work-life balance,
perks, like other responsibilities?
So at Microsoft,
like things and responsibilities
beyond when you're like on call,
it was measured on a weekly basis.
Like, you know,
you meet your manager on a weekly basis
and it's measured if you're like exceeding,
meeting or like not meeting the expectations
the manager has set out.
But at a startup,
it's not measured on a weekly basis.
It's measured on a daily basis.
So like,
if like there are goals you discuss in the morning
that, hey, you need to do these like five things
or sometimes 10 things or three things,
you have to like push yourself to meet them.
And there's more pressure to achieving them
because you have a lot more ownership
and you pretty much own end-to-end
of whatever you're working on.
For example,
let's say I'm owning a hackathon
that I'm organizing this weekend.
Like it's called Build Matcha Build Code.
Like you build your own matcha,
build your own code.
I own like promoting,
marketing,
and like planning the hackathon,
being innovative with it,
choosing the companies to partner with,
like having constant meetings with them,
figuring out the supplies
and all the dynamics you have to own
and do it.
Doing like all the hacks.
Yeah.
And now since like all the pressure is on me
to have it possible this Saturday,
I have to like work.
Not working is not an option
because deadline is the Saturday.
So the pressure is higher,
but since it's more fulfilling work
because like,
I think I was like a bit fed up of engineering
and that's why I enjoy it way more.
I'm having a more fulfilling day,
I'll say.
Wow.
Yeah, that's definitely a change,
but a good one.
100%.
You switched jobs
in one of the toughest job markets we've seen
and you moved to Silicon Valley.
How hard was it to do so?
I thought it would be easier,
but I didn't know that it's extremely hard
to land a job in Silicon Valley.
I thought like Silicon Valley
is the land of opportunities.
It'll be extremely easy,
but it was absolutely opposite.
It was extremely hard.
I left Microsoft like eight months ago.
I moved here.
I kept on applying online
to a bunch of companies.
I applied to 200 companies,
got 10 interviews scheduled
and all 10 rejected.
And most of the reason
that I got rejected
was that I applied first
for engineering roles
and I figured out that,
oh, by doing 100 LeetCode problems today,
I would be able to ace them.
But no,
the problems they ask
are very unstandardized.
They're not standardized.
So think about Big Tech
when you applied for Microsoft, Google,
like Meta, etc.
Interview process is standardized.
Like, you know,
they're going to ask you
like LeetCode hard, medium.
But when you apply for a startup,
they can ask you anything.
They can ask you HashMap collision strategy,
which I had completely forgotten.
I studied in college
and now completely forgot about it.
Or they can ask you about
or do a system design of Bitly
and do all the mathematics
that's needed
so there are no collisions
in that like, you know,
bunch of links you create.
So these tough problems
require a lot of in-depth knowledge.
And I realized
that I needed like
at least one month
more to prepare.
Like reading the system design book,
doing 100 more mock interviews
because the intensity
of the job market
is at the peak.
I see Reddit blogs every day
that people are struggling
to find jobs
even eight months
after trying and interviewing.
Then I realized,
hold on,
let's try DevRel.
Maybe engineering is not for me.
I might
might
think about engineering later.
So then
with DevRel,
you don't need to do engineering.
You need to know
engineering basics
but not need to do it.
And
that's why
I pursued it
and got a work trial,
not an interview
and got it pretty easily.
And I was considering
another DevRel
company.
I won't name it here.
It was a big tech.
Right now,
I wanted to work for a startup
so I was just
prioritizing startups.
Yeah, I mean,
it seems like DevRel
is like the perfect fit for you.
Both.
You get like a little bit of
development experience you need
and a little bit of like,
you know,
marketing experience
and it's really fun.
It's a good hybrid approach.
Yeah,
that's so fun.
Speaking of a tough job market,
do you have any thoughts
on the recent layoffs
from companies like
Meta and Oracle?
So,
this is really interesting
what's happening.
Big banks,
they made
a bank
in the last quarter.
In January 2026,
they said
AI is not going to affect
your job market.
And by April,
they announced
10,000 layoffs.
Most of the big banks.
And what's happening is
we are living in an
inflection point
after Cloud Opus 4.5.
Now,
Cloud Opus 4.7 is there.
That inflection point
is so big
that,
like most of the engineers
I know in Silicon Valley,
including Spotify,
you talk about in your videos,
that
they haven't been
doing code with hand
for the last
four months,
like even five months now.
Since November to now,
like no one is coding with hand.
They're using
cloud code or
agentic terminals
to code.
So because of that,
I think the productivity
has skyrocketed
and
engineers
who have
nailed down
their syntax abilities
like TypeScript,
we have become
very language agnostic.
Think about an engineer
who just learned Python,
they know how to
reverse his list
or they know
a basic for loop
or think about an engineer
who did the same
in TypeScript
or web development
or changing a button
who spent years
on perfecting that skill
that's completely gone.
So
the core skill now
which is more important
is engineering,
scaling big systems,
scaling from
1,000 users
to 10,000 users
so your system is
running on
cloud platforms
or bare metal.
These questions are
more important than before.
So people who do not
have enough experience
with like
scaling,
cloud engineering
are struggling the most
because
when I went to college
even 2016-2020,
scaling and system design
and these intensive systems
were not even introduced
but now
colleges are being introduced
so people who have
like
older college skills
but did not get
the bright work experience
now
are struggling the most.
So
that's why
people need to
nail down their
system development
system design skills
way more than
coding skills.
That's what I think.
What are your thoughts?
I 100% agree.
I think as
AI basically makes
knowing syntax
and knowing
leet code
easy, medium, hard
more obsolete
it becomes
more and more
important to
know the big picture
know how
components fit together
because you can write
any AI can reverse
a linked list
or balance
a binary tree
but I feel like
sometimes
if you don't prompt
or if you don't
know the system
that you're building
you will end up
with a solution
that technically works
for one use case
but then completely
falls apart
as you scale
or as you add
more components in.
100%
You mentioned
that you specifically
were looking for
startups to join
right?
Do you think
that startups
adopt AI quicker
than big companies
or was there
a particular reason
why you really
want to join a startup?
First of all
I'm working on
a very small startup
it's like a seed
they have just
raised a seed round
so startups
work differently
as if like
they don't have
much bureaucracy
first of all
and other reason is
that they use
the latest, greatest
tools that I wanted
to try.
For example
I will give an example
like Linear
we use not
Atlassian
it's a new product
I'm getting to try
Superhuman we use
not Gmail or Outlook
so we use
Cloud Code
not VS Code
so all these
experiences
and getting people
to know what are
these latest, greatest
tools doing
and are they worth
trying?
This is not
just an experience
to work for a
smaller company
but also to try
different tools
and how they work
and how they integrate
and to figure out
AI is actually
adding value
to our lives or not
so for example
in our startup
we have installed
an OpenClaw instance
we are running
an OpenClaw instance
it was before
OpenClaw
now it's Hermes
so with Hermes
we can ask
Hermes
give me access
to this GitHub repo
I don't have access
I just joined
I'll just ask
to that AI chatbot
it'll give me access
so that OpenClaw
Mac Mini is really
powerful and I'm able
to get use of it
so trying that
Mac Mini
setting up that
Mac Mini
which we read
on Twitter
and actually
implementing for
your startup
is really fun
nice
you would never
get that experience
in any big tech
yeah
I'm also just curious
so you mentioned
OpenClaw
I know it's a really
really powerful tool
but also
you have to be very
cautious about security
how does your startup
make it so
someone random
can't hack
your security credentials
or things like that
security is a big
concern
you're right
but for that
we're using NemoClaw
we use NemoClaw
because that was
with guardrails
from OpenClaw
and with Hermes
we added
we tried to
add as many
guardrails as possible
for example
it never gives admin
access by default
it only gives read
only access
it can onboard you
pretty much to 90%
but the remaining 10%
we have to do
through humans
so it's a pros and cons
if you give
like 100% access
to GitHub
to a new hire
they can literally
make new
car GitHub repos
so you cannot
you cannot trust
right
so that's how we prevent
yeah
guardrails are so important
I feel like
yeah
there's definitely
been like
the horror stories
of like
someone
some AI
deleting
some like
production database
or like
the meta head
of some department
that like
used OpenClaw
accidentally started
deleting all her emails
yeah
that was wild
yeah so definitely
use AI
but guardrails
guardrails
100%
like I mean
there are so many
startups too
including Context AI
Warsaw
like they became
victim of
security issues
oh yeah
that wasn't even
their fault
it actually was
wasn't it like
they were using
a company who
got SOC2 certified
using Dell
yeah
we know how that went
100%
and then someone
someone
downloaded
robots cheat codes
onto their
corp laptops
so that's how
it got hacked
oh my god
so not only
do you have to
protect your own
you also have to
be very cautious
of the software
that you're using
100%
so the point
I was trying to make
is that
that all happened
through social
engineering
so like you can
get an employee
like trigger them
in one way
and that can
really happen
so for example
today it happened
someone from a
new number
tells me that
hi I'm your
CEO
with name
I am like
name of my
CEO
it felt very real
and then I
showed this
number to my
CEO
is this your
number
and he said
no
so social engineering
is the way
people can trick you
oh yeah
100%
and especially
I know like
you know with
the ability to
clone voices
and to
spoof stuff
it's so much
harder to actually
tell if someone
is who they
say they are
online
100%
let's just do
some rapid fire
questions now
just say the
first thing that
comes to your
mind
Microsoft
Microsoft or
Hydra TV
oh
I live in
present so
Hydra TV
Leak code or
real world projects
will AI replace
engineers
yes or no
no
no okay
one skill every
developer should
learn in 2026
system design
100% agree
most overrated
thing in tech
right now
salary
most underrated
thing in tech
right now
learning
high salary or
fast learning
learning early
career
you need both
but I think
it depends right
like if you
want to pay your
debt if you
like let's say
you had a
college debt
you would choose
high salary
if you don't
have a
college debt
then you
will choose
a high
learning team
so it
depends
got it
job hopping
or staying
long term
job hopping
I mean think
about it we
live in a world
that even
open AI
co-founders
were not
like you know
were not
loyal to their
own startup
they found
they left it
XCI
they founded
XCI
and they
left it
so if
co-founders
are not
loyal to
their
startups
how can
an employee
be loyal to
the company
they work
it's such a
competitive market
yeah yeah
content creation
or coding
content creation
Devrel
what's one
mistake 90%
of new guards
make
that would be
not increasing
the surface
area of luck
just hiding
in their
room and
applying online
I think going
out there
meeting people
is the
biggest way
they can
like be
public
plus
having
Twitter and
LinkedIn
people who are
not on
Twitter
they are
actually facing
the bigger
loss of
not being
exposed to
a job market
which is
only for
Twitter
users
same with
LinkedIn
same with
like
Substack
same with
Discord
I know
people who
found their
jobs on
Discord
Twitter and
LinkedIn
you gotta be on
all three platforms
in the right
communities at
the right time
I've never gotten
my job just
called applying
online so
totally totally
feel like
not just
be a good
engineer also
having those
access to
the communities
100%
so two
things would
be like
IRL
in-person
events and
number two
would be
finally if you
were starting
from scratch
today what is
the first thing
you would do?
I would go
out go to
tech events
and see what's
the hottest
discussion
happening right
now for
example in
Silicon Valley
the hottest
discussion right
now is
continuous
learning by
agents like
I went to
today a
store called
Ando Market
that store is
completely run
by AI
the biggest
problem in that
store is it
doesn't learn
as you go
because it
doesn't have
contextual memory
this is a
very important
research problem
lot of people
working on it
I would work
on that or
try to get
my experience
in one of the
hottest problems
and try to
build my
experience and
work for
startups doing
that and
show my
passion
And one
final question
is coding worth
it in 2026?
Coding
I think
learning to
code is still
worth it but
just learning to
code is 100%
not worth it
coding is pretty
much solved but
still we have to
learn like it's
2015 you have to
learn from scratch
you still need to
like know the
math you still
need to know how
to calculate 2x3
even though you
have calculator
you can use a
calculator but
learn like you
do not have a
calculator
So I think more
you learn like it's
2015 you don't
have AI and
work like you
have AI that
should be the
way you should
work.
My friend Kier
he taught me
when he was
here
So shout out
to him
Cool
Thank you so
much for your
words of insight
and wisdom
this was amazing
we learned so
much from you
if you have any
questions for
Hardware put them
in the comments
and we have a
video out on
your channel so
definitely check
that out as well
Thanks for watching
and see you next
one
Thank you
Yay!