Amazon - Aptitude & Interview Q&A-

20230917_160435-Meeting Recording
September 17, 2023, 10:34AM
1h 6m 8s

Speaker 1 0:06
Hello.

Speaker 2 0:07
Yeah.
Yeah, you are.

Speaker 7 left the meeting

Speaker 1 0:12
Ohh hello everyone. Good evening and welcome to the island. Nice
afternoon, Amazon. Today we are joined by Alien Nai Atherley, Parag
Vishwa Ritik and Shahid, who will be sharing with us their internship
experience and will be also talking about that project and will also help us
clear our doubts regarding aptitude and interview. So the doubt the
wasting anymore time, I would like a 30 to take over and talk about his
experience at Amazon.

Speaker 8 joined the meeting

Speaker 9 left the meeting

Speaker 10 joined the meeting

Speaker 3 0:43
Yeah. So I'll start. Hello, everyone. So starting about my internship
internship experience. So basically the majority work at Amazon is done
on SQL, SQL. So you are RDBMS which is our RDBMS data, relational
database management system. So first the first of all thing that you, you,
everyone should be like completely advanced in SQL.
The query languages, if you are applying for the business analyst position
check will tell you more in detail about the business, research analyst
position.
For the business analyst position, the majority work is done on sequel, so
like Lev. Joins, right joins, then data normalization and other different
kinds of data.

Speaker 7 joined the meeting

Speaker 3 1:31
It is majority work on SQL then there is there is few parts on dashboarding
which they'll teach you how to do the dashboarding and stuff and in very
rare cases you will get to do modelling in BA work. So in my case I have
got few opportunities for modelling work as well because my my team had
few models which were like poorly performing so the at my internship the
3 main parts where first was to design a metric.
Ohh accuracy, my trick to define how is the performance of various
models and how are they working the second. The second thing. The
second thing was 2 to make 2 plan things that how can we make this
model if any model is performing badly then take that model and do
feature selection and all that stuff to make sure that the model is
performing better and 3rd one is complete end to end documentation of
this.

Speaker 11 joined the meeting

Speaker 12 left the meeting

Speaker 3 2:31
So in the first part of seeing how the current models are performing, we
have to write SQL queries to pull the database to pull the data from the
database which has both the actuals and the predicted values. So the
predicted values you'll get from a different database and by writing the
SQL query and you'll also get the actuals from the database by writing a
different query. So see it is completely done on SQL and after pulling both
the data you have to like.

Speaker 12 joined the meeting

Speaker 3 3:01
Use your accuracy metric to get the final like output on the what is the
accuracy? You should be also more inverse with advanced Excel like we
look up and all that stuff because that is like on preliminary level and then
that was the first part. The second major part was like after pulling the
data I saw that there was there was one specific region which was Japan
region which had poorly performing models. So the second step was to.

Speaker 12 left the meeting

Speaker 13 joined the meeting

Speaker 12 joined the meeting

Speaker 3 3:33
Like completely analysing why that specific model is not performing well
like it may be due to inappropriate feature selection and also testing that
whether the model is performing poorly now or it has been performing
poorly since it's whenever it was brought into production. So the JP model
was brought into production 2 years ago. So at that time the accurate
accuracy was good so the accuracy fell down with time. So that was the
second part.

Speaker 14 left the meeting

Speaker 3 4:04
And then we moving ahead. There was also you have to do a complete
end to end analysis like if you have found out that a specific model is not
performing well, you have to also find out why it is not performing well. So
after that we have to also give.

Speaker 11 left the meeting

Speaker 3 4:20
Ohh like solutions to your team that what can you do or what can the
team do to make sure that the model does better predictions. You have to
also understand the concept of computing efficiency. So there are few
models that give you very good accuracy, but they are computationally
expensive like xgboost, random forest and all that stuff. Whereas there
are some cheaper models that give you decent accuracy like logistic
regression and with appropriate features selection you could also get
better accuracy.
Such models, so computing efficiency is one of the things that they see at
Amazon, like whatever you compute, there is a cost A tried like cost
assisted to it. If you're computing for $100, your model must be giving
you profit of at least $150 or more, and so on. And the 3rd part was
modelling itself. So I'll also like to highlight that access in Amazon is very
limited. You won't get access to everything from the start.

Speaker 15 joined the meeting

Speaker 16 joined the meeting

Speaker 3 5:20
You have to work your way through a way out through it and as
everybody's expecting that they'll get ML prediction models only. So don't
keep that expectation. That is completely like on your team, some in
some team, even DAS get work to work for the predictions and all, but not
every team has that.
Like there's modelling, so the majority work of be a what I understand
from Amazon is like SQL work, then calculating the accuracies, doing
analysis. Why specific accuracies coming lesser if they. If there is any like
losses in any department, why is that loss going on assessment of that
and dashboarding so that is the major work at Amazon and talking about
the interview you have to be like they'll ask you to code various SQL
situations.

Speaker 17 left the meeting

Speaker 18 joined the meeting

Speaker 19 joined the meeting

Speaker 20 joined the meeting

Speaker 3 6:09
In the interview itself, by sharing your screen, they'll ask you basic
theoretical SQL questions like what is primary key, what is secondary key,
what is normalization in SQL and so on.
Then.
Ohh, moving ahead there also give you situational questions that what
you will do in specific questions like in specific situations like they'll give
you a call. Let scenario. They'll ask what difficulties you have faced in your
internships like which is the summer internship and how you overcome
that specific problem. Those are the situational questions. So there are 2
main rounds of interview in the first round of interview it will be different
for everyone like some might get.
Or technical in technical questions in first and behavioral questions in
second. While some might get wise verse and some may not even get
technical questions. But for me the first round was technical questions
where they asked me about general applications of Tableau that I had
done in my previous internship. They asked me about Power Bi. Then they
asked in detail about my projects. The feature selection I've done, what
problems I came across the project and how did I find a solution. Then
they asked me.

Speaker 21 joined the meeting

Speaker 22 joined the meeting

Speaker 3 7:25
General question about Python that what is list? What is a double and also
asked me the syntax for the same. Going towards SQL they asked me
what is data normalization.
Then what is left join? What is right join, inner join and a few they have
given me an example that this XYZ is the input table and they want a
specific output from that table. So which one should be done? And please
give the code for that so they asked me to share the screen and type the
code for SQL and also if the for one specific problem there are multiple
approaches that.
Like you can do it through joins, table joins as well as through.
Case there is also a case when and there is also one more technique, so
subqueries. So there are multiple ways to solve one problem. You can give
anyone solution but when you give the solution you can also mention to
the interviewer that there are various other approaches to this as well like
this and this and moving ahead from that. This was my first round the
second round of me was taken by non technical person. He was AC CA so
he had generally asked me behavioral questions. So as I'm in.

Speaker 23 left the meeting

Speaker 3 8:44
Basement cell? Yeah. Ask me that. What? Different work. There's a
placement cell carry carries out. What problems? We are faced in the
placement cell. And how did you overcome it? And you also asked me
about my internship. That what was the problem? And do you face any
problem? And how did what approach you did to our commit. And he had
given one situation himself that how will you overcome this situation?
Please give me a detailed approach or a standard operating procedure for
the CM and this format 2 rounds of interviews.
And about my internship completely summed up.
If anybody has any questions you can ask.
Hello.

Speaker 1 9:42
Ohh cancel. Other alumni also shared their experiences. Similar matters
spiders mean time others would post.

Speaker 3 9:48
Yeah.
Works, works.

Speaker 4 9:52
Amisha gonna go next.

Speaker 1 9:52
Barack was you make call.
Language.

Speaker 4 9:56
Hi. Can you all hear me?

Speaker 1 10:01
Yes.
Speaker 4 10:02
Ohh.
I am to heck and I interned at Amazon as a Business Research analyst.
So let us start first with aptitude. How will your aptitude look like? Your
aptitude test will be of 90 minutes and it will be monitored.
The first section will be logical reasoning questions and aptitude questions
in general, but the thing that they do is they give you less time in the first
round, wherein you have to answer 25 questions or so in 5 minutes.
So I would really recommend all of you guys if you are not practicing your
aptitude, please do so even when it comes to shapes etcetera. Because if
you leave one or 2 questions then you might just get afraid in the next
few hours with because you will think he Mirai, Miss Okay and will do bad
in the further sections.
So really, really practice or aptitude very nicely.
Though good part is, they do not ask much complicated questions to you
when it comes to machine learning or hypothesis related questions in the
aptitude test itself, most of the questions will be on the basic
understanding of concepts like accuracy, precision and what is what is in
null hypothesis, what is a type, one error type 2 error, P value? What is
the significance of P value and these type of questions and some other
simple probability questions from base theorem and just computing
probability of X by with given a table.
Having probability distributions.
It's very simple, but the tricky part comes wherein you have a section
dedicated to.
Ohh, data warehousing and ETL.
So for specially for the BA people, e TL is very very important because
that is what Amazon also primarily uses, which is to pull the queries by
extracting, transforming and loading the data inside SQL. So your SQL
needs to be very, very well versed because there are SQL questions in the
aptitude round as well. And there are pseudo code questions which are
based on similar logic as what is followed in SQL. Like Arthur mentioned
that you should be very well versed with how subqueries inside of how
you can search for a text etcetera should work.

Speaker 24 joined the meeting

Speaker 25 joined the meeting

Speaker 4 12:10
Because there are very less Python questions in the paper of aptitude, but
there are pample of SQL and SQL type of questions, your logic behind
loops should be very very clear, because whatever logical reasoning or
something question they'll give you will be something very very similar to
loop itself. Then I don't remember if there is a section related to English or
not, but there is a dedicated section towards data leaking. Warehousing is
well.
For which you will have to refer to the Amazon Curve website of a WS
only. They explain what is a data lake DOT data warehouse, what is
software as a service, etcetera. So these will be questions which will be
asked. We also did not have much about that when we gave our test
because that was just a 35 words in the whole job description. But they do
ask that.
Ohh.
This this, what the aptitude will look like.

Speaker 26 joined the meeting

Speaker 4 13:09
To position you will be allotted will be based on your aptitude score and
Amazon will internally ohh assign positions to you. It does not matter what
reference you choose.
Now the interview I for Ari interviews.
Every single word on your CV will be asked from you. Whatever you have
put in, just put in what you can explain very well. Do not put in
exaggerated stuff which you cannot explain at all. Don't put in that you
know tablet if you're not even fat familiar with Tableau. If you just know
power, be it just put. Power BI don't lie on your resume in the most
simplest words I can say this because they will pick that and ask you
questions from it.

Speaker 27 joined the meeting

Speaker 4 13:52
Your academic projects will be of significance to them, especially if they
are done on textured related or.
Recommendation system. They are very interested in that. I also highly
suggest that you read about how Amazon uses NLP NER in their
businesses. You can find plenty of examples online as well. We'll also give
you examples on what we work through and what type of coding they do
there.
So this is what your first interview will look like up supposingly if you have
done a project on recommendation system of books. Also they will ask you
what kind of algorithm did they use. Do you did you encounter anything
difficult during that and how you dealt with that problem? Because in our
interviews most of the cases we did not get asked very technical
questions as such of statistics or even coding. But they also answered how
we were behaving towards or approaching a problem when we
encountered key.
An obstruction or like hurdle sort of. How did you deal with it? Are you
afraid to ask for help or not? These are something that they will be very,
very looking forward to what your answer is there, because that's how the
short listing also takes place.
Then comes down to the projects that you are putting we.

Speaker 28 joined the meeting

Speaker 4 15:10
I would highly suggest that if you have text projects, even if they're not
done in college as a part of your academic project, mainly if you have
done it, I'll say please put it on your CV. It gives you a major point when
you are interviewing at Amazon and before your interviews take place.
There is a smart technique. I'll mention what smart means in the chat for
you guys to understand that that it's how you can breakdown your entire
project into 5 parameters of.
SASMERT, which is specific, measurable, actionable result and time
bound. How much time did you receive in what was a specific objective?
What did you do to attain that objective and how much time did it take
you? So you have to boil down all your projects into those categories and
then explain your projects when you are doing so in the interview it will
reveal a lot of edge over other candidates as well and we'll be something
that they also value a lot.
Inside Amazon, when we are also.
Given to fill out our sheets, they also ask US smart ideology. Perky. How
did you do this and how did you do that?
Other thing for interviews or go through Amazon's 17 leadership
principles. What they mean by that? Because they'll ask you what really
leadership principle do you find yourself most inclined towards, or what is
something that you already follow etcetera. So you can just note it down
by before you give the interview or just put the list near you to select and
explain why you resonate with that leadership principle. If you are saying
that you think out of the box, then you have to mention.
Open leadership principle, similar to that that, yeah. Amazon has that.

Speaker 24 left the meeting

Speaker 4 16:55
Now the second round of interview is where the non tech can also take
your interview and in some cases they are somebody from where you will
be joining might be taking you an interview itself. For me the guy who
took my interview was AB 8.
And he told me a bunch about how he's.
Ohh, his day looked like when I asked him so and then they asked me
what I had done in previous internships. Even when they were not
statistics related they will ask you every question if you have put in
volunteering experience, Co curricular activities they'll.
Ask you what did you learn from that? Why did you do it, and how did you
attain goals? If you were in a leadership position there? Or what goals did
you set for yourself, etcetera. And how do you generally deal with
rejection when it comes to something that does not work out your way,
etcetera, and how often do you ask for help?

Speaker 24 joined the meeting

Speaker 4 17:50
So these are a lot of things that they focus on, mainly leadership
principles, because Bhat in the most literal sense. There are people like us
who prepared a lot for all this and there were people who just got it on the
basis of leadership principles who were not worked with Python as well,
who were in these internships.
So Please remember that that they'll hire you on how you are interested in
their company.
Not just based on your skills.
So this is all my projected Amazon was.
I worked on Transformers language modelling for correction of some
browse node and mapping throughout a little complicated to explain in
the end I ended up using a semantic similarity which is how you can make
sense of the word or the sentence and assign them to another, something
which is similar to them. Like if I'm talking about coffee maker then I'll
assign it on the semantic basis to a kitchen category. So that is what I
worked on, I worked.

Speaker 29 joined the meeting

Speaker 4 18:54
Most arrays work majorly on Python And SQL is a part of the curriculum
that all arrays and bias are supposed to go through, and for pulling your
data your SQL should be good because at cases you'll not get your own
data, you'll have to pull it and for that SQL is very necessary. Don't worry,
they will give you a good one month cut training essay and you will have
to attend all those trainings in sessions. They'll explaining everything what
they do, how they do it. There are also some Python classes included in
the.
Cleaning curriculum. Happy. So that is there. And other than that, other
skills that you need as an array or for your position as a data centers that
please be well versed with how to treat raw data?
Ohh if they ask you in the question if what is a data quality, how do you
govern data quality or?
Keep missing values. What is the best way to fill it etcetera and
everything else keep mention that it depends on the use case and the
person who will be the stakeholder of the same who is going to be the end
user. So that's how you should be framing your questions and.
One important thing, if in case your interview ask you, do you have any
questions for them? I highly I would highly appreciate if any of you most of
you ask them a question because that would show them that you are
genuinely very interested.

Speaker 30 left the meeting

Speaker 4 20:23
If anybody has any questions please.

Speaker 2 20:31
Hi Jack. Ohh you said in your.
On your statement that.
Ohh based on your aptitude test, no, the rules would be assigned and not
to preference that you gave would be considered. Is that correct?

Speaker 4 20:47
Yes.

Speaker 2 20:49
So what sections in the aptitude would give you certain roles? Can you?

Speaker 4 20:55
Actually that is something. So I did ask my colleagues when people who
created the test when we were. I was also at Amazon and they said that
it's assigned by us to some internal parameters that is not revealed very
well to any of us.

Speaker 2 20:55
Daily Star.

Speaker 4 21:09
So yeah, maybe it depends on the speed at which you are answering how
fast you're completing, or how much time you're taking over a question
because probably they don't monitor screen time Kitna, they do question
pig etcetera. But as far as we know, nobody knows exactly he will kiss
basis person. Karthik wants a section key basis person Karta.
But there is this because uh hamara most of like, I also opted for B when I
filled out Amazon. But when my list came after the aptitude test, I was
automatically assigned for the array interview. It wasn't that I had chosen
it. My first preference was to be a profile, but due to the score, a lot of
people had this happen to them. You know, they got a different profile.

Speaker 2 21:56
Okay, thank you.

Speaker 4 22:00
Poor preparation. Uh, you guys. Yeah. Sources you guys use for SQL.
There are a couple of them. If you are not very well versed and would like
a quick refresher, I highly suggest using data camp to do so. It has a ton
of great videos for you to understand this thing and secondly ACHA. So
are you guys aware that SQL has different types of SQL itself?
You can just raise your hands or something. Send this key you know or not
that part.
Right, so SQL is different types, so be well versed key sentex sometimes
changes a bit according to the same, so just be well versed because your
interview might ask. It only concerns SQL ATA Hai if you know mssql, say
mssql. If you know my SQL say MySQL, don't say that SQL is same for
every place. No it's not. SQL differs from place to place in Amazon the
user specific SQL called Red Ship SQL. So it is different from what we use
normally.

Speaker 31 joined the meeting

Speaker 4 23:05
Which is amnesia squalor miniscule, columnist of caps me like the
command Kayu myself shot me like me. Can you make command, Elsa,
and may, before each sentence break. But you mean to those are many
small, small differences that tumhe Pata because your interview might
interview my task.
For level of SQL please prepare to the advanced level of SQL. I'll drop the
resources that I use for practicing lead code for basic practices and hacker
hacker rank is also quite good for the same and data camp for videos
etcetera.
For quick revision, I'm also mentioning some sites that Ohh which were
yeah it's OK if you know any SQL. Basically you should know how an SQL
query works. Key Augur like a command error. You should know Agar to
me. If supposingly Bola extract all the words in one column which starts or
ends from a or it has as in it, then you should know how to write a
command because within different types of SQL query the commands that
may change.

Speaker 17 joined the meeting

Speaker 4 24:16
You know that, but you should understand that logic very well.
Any other questions?
You can also go through with geeks for geeks and W3 school for your
Python division as well. Python will be very basic asked, but Agar Tum
name.
Project manager mentioned Kiya Hai.
Please what Chasse Padikano could become?
Facebook.
And don't say but I wanna check it. I know sometimes they will put you
ask you multiple multiple questions on one project only. Sometimes they'll
go through everything. Kavi Kuch puchenge depends totally on the
interviewer.

Speaker 32 joined the meeting

Speaker 4 25:14
And do another tip that I like to give you all if anybody knows any neural
network.
This thing.
Mention deep learning as one of the skills in your CV.
It might make only a tiny difference, but it's a huge thing. If you know that
already.
Ohh so.
Take a lot of questions, ABI.
If then I'll be project is mentioned in your resume, will they select you for
RA totally depends on the interviewer and how you're giving the interview.
Some of us did not have any NLP projects and were selected for the role
does not depend.
And sum up things that we need to know about Amazon.
That would include leadership, principle and mostly how what are their
businesses mainly if like Alexa concert division may have something put
about cheese.
Not only to you to make you headquarter copy here cup start weather
India may kabaya and with what are the major products that they sell a
W.
Just came out, but I wanna tell you what is a WS and what are the other
products except for Amazon Shopping and Amazon.
God web services that are building them business, they will not ask you,
but it's good to know.
Hybrid and working person. I think it's hybrid now for everybody. It's not
going to be virtual unless your team is based out of India, then it will be
completely remote rule.
You can go to the office, though still with the ID card. They will allow you
to get an ID card that you can go and visit the campuses.
And I think for recap on the questions asked in app, I think author will,
Parag and Vishwa will also be able to help you on the same. I don't
remember most of the questions, but they might do so. And for the
negative marking as well, I am not aware if there was any they might
know.
But AG, Vishwa, can you please ohh answer these?
Ah, for neural network projects. Just a second or 3rd. Why it is answer this
question.

Speaker 3 27:52
So.
Yeah, yeah.

Speaker 4 27:57
Ohh, for neural network project. So I mentioned on my resume that I had
made a chat bot. So for chat bot they'll ask you what kind of what kind of
processor that you follow. So you should know the pipeline of how a
natural language.
Natural Language Cup pipeline case worker how do you convert the data
set key parallel lower case supposingly tumhari sentence it to first school
lower case may convert karoge punctuations. Remove stop word saying
and remove car okay and you might be able to understand this very good
when you are taking the Class 4 naive days and in especially for assess
students in the imlt course a lot of this is already included and you'll get
to know how the pipeline of this works.
That to person, whoever is teaching you will be able to explain you this
very, very well, key stopwatch, skewed, remove curtain etcetera. It's for
improving accuracy later because those are not so important, but highly
appreciate that you please do a quick refresher on the same.
How often? So yeah, that's there. If you have anymore questions, please
drop it. In this chat, I'll answer on the chat itself. I'll just answer this one.
How often.
You asked for it so.
Angela saw The thing is, they want to know if in case you are stuck at
1:00 point because will you are you willing to ask for help or not?
And are you able to accept the help given to you or not because.
What will happen is that key at times you will not be able to understand
what you have, what you are doing or you might be at a very difficult part
in your project where you're not able to comprehend key, what to do next.
Because this happens with a lot of US, key data is not good, something is
not coming up because even I tried like 7 different models for the project
that I made and nothing was working out. But even in the interview when
they asked me if you get stuck, what do you do next? Like I answered
that.
I asked my fat friends who are familiar with these fields to help me out. If
not my friends in case my friends are not capable enough to answer me
the same, I will ask my seniors or so all my teachers in college if I'm
encountering a problem.
So they just want to see how do you respond to?
Ohh dysis he Agar kuchh Jada. Are you willing to help on willing to ask for
help? Are you want to do you want to keep working on the same thing or
do you want to just give up if it gets stuck at 1:00 place. So that's it. So
that is there.

Speaker 33 joined the meeting

Speaker 4 30:39
Ohh I thought. Please can you answer the other questions.

Speaker 3 30:43
Ohh yeah sure about Ms Excel.
In case of Ms Excel you will have to like no advance Ms Excel like there is
even dashboarding in Ms Excel and we look up and all that stuff.
Other than that, I don't think so. Weird. Use Excel much, but yeah for that
for making dash like small scale dashboards and other friends.
And other than that, yeah, about the aptitude, I think so. The aptitude isn't
the same for every batch. The aptitude and it's questions will keep
changing as check told you have to go through their website like read
about cloud and all that stuff, read their principles and all that stuff. And
you should know like in case of machine learning questions and all that
it's like very tough because the options are very close to each other and
they'll give you a scenario in which we have to tell which.
Ohh algorithm you will select and like you know you should know
applications of it, not just the theoretical part.
Then.

Speaker 12 left the meeting

Speaker 3 31:50
Will they ask questions? If. Yeah. Yeah. So if you put certifications, they'll
get definitely ask you questions on that. I'd put certification of Tableau in
my CV and they had grilled me a lot on that. So if you're putting anything
on your CV.

Speaker 4 32:03
They will grill you on anything that you put another Siri.

Speaker 3 32:05
Yeah.
Anything you put in your CV, you should think 10 times before putting it
on your CV.

Speaker 4 32:11
Yeah, they can even ask you to open your camera, your screen, and share
your screen in code. Some logo Kafi logo.

Speaker 3 32:15
Yeah.

Speaker 4 32:19
Be here. May I was lucky. Camera camera be opening Harada interview.
She was very charming, interviewer Mirai. See interview. She only asked
me one statistics 2 stats questions. The rest of the entire time she was
asking about my projects.

Speaker 3 32:36
And you are asking if it is fine if you know only mystically it is fine. You
should only be well versed with the syntaxes. My skill is not a problem.
They won't ask you the basic, they will just ask you the basic questions,
not at very advanced level though you should know the advanced level,
but they'll ask in if they ask you to code in the interview, they'll ask you at
the basic level like left join, center, join, right join, type of questions.
Like for example.

Speaker 4 33:03
Exactly. And the same goes for Python 2.

Speaker 3 33:06
Yeah, in Python they'll ask you what is a list and everything. Yeah, loop
still ask and.

Speaker 4 33:07
He will looks painful to you.
Yeah, yeah.
Loops. They left initially ask.

Speaker 3 33:13
Hmm. And what is like, what is the difference between a topoil and all that
tuple list? What is the difference between that such questions they'll ask.
But to be honest, if you are shortlisted for BA, you won't need much of
Python because you won't even get access to Python in the start of the
internship.
So we make sure you're escale is on point. And yeah, whatever you put on
your CV, it's on point.

Speaker 4 33:44
And please, please do not exaggerate on your CV.

Speaker 3 33:44
Hello.
Yeah, that's true.

Speaker 4 33:49
Very kind request to all of you. Please don't exaggerate at all if you have
not completed a certification. Please don't put it because they will ask you
questions on the same.
Speaker 3 34:12
And about hybrid or in person, somebody's asking. So you have for most
of the team it is hybrid for now and it is 3 days a week for now.
Like 3 days you have to go to the office and for the other 2 days you can
work from PG not home. You have to be in whichever location they tell
you.

Speaker 4 34:47
Anymore questions guys?
Acha underrated please other questions. Take me to Amazon.
You put money to you. Took him some.
Tumhari Field coffee software developer karlena aptitude now across all
tech and non tech profiles.

Speaker 3 35:10
And don't be much tensed about the aptitude because this tell shortlist
the complete class. In our case I think so. Check from your class. Also
everybody was shortlisted, right?

Speaker 4 35:20
Haha. But like the first?

Speaker 3 35:21
What then?

Speaker 4 35:23
Think about the show mother a good 52 shortlist with him anytime to first
round of 26 candidates were there.

Speaker 3 35:30
Yeah, right.

Speaker 4 35:30
Total of 2 profiles and they cut down a lot, lot and B method. They hire a
lot of people, data scientists. They are hiring for the first time but I don't
think completely hiring who gives.

Speaker 3 35:32
Yeah.

Speaker 4 35:43
Because we internally botlhoko there upskirt.

Speaker 3 35:48
Yeah.
And also for the data scientist role, the inclination criteria is more tough.

Speaker 4 35:55
Hot like hot. This is true because they will mention that you need to. You
will be building advanced ML models.
Who with Nigel?
Read it now.
Taking me later, I got.

Speaker 3 36:06
I don't.
Yes, true that.

Speaker 4 36:09
Because they want you to know complete deep learning and as
complicated as your project is, they want you to be able to explain it very
well. So you are a galti mat Karna.
Because if you were selected.
That to complete.
Just got back. Give again.
Manipur Chaki LSTM 128 units or 64326 units make up for a code that
Mahaffey Galti mat cardiac please.
We might ask you if you put s neural network connection is a poor
network exactly work.
To the consulate is labelled roommate keep. Please do not lie on your
resume, they will pick and ask every single thing possible.

Speaker 3 36:50
And also the tougher position you will apply for getting inclination vote for
that will be even like more tough on the same line, like if you're applying
for TDs, the I like I had the when I was working at Amazon, my manager at
shared me the inclination criteria. So in that it was given that for data
scientists, the inclination, the inclination process was a lot tougher than as
compared to PA.

Speaker 4 37:21
Because I don't know bees. May Hota any hotel Aries, major, when I was,
we were in turning to after project got completed. Review calendar. We
have to submit our 2520 pager documents something which includes all
your documentation about your code etcetera results and everything with
Appendix appendix support.

Speaker 3 37:22
And.
Yeah.
Yeah. True that that I had also got to submit the same. And moreover it's
like for our prior batches that is the batches before us, the getting an
inclination vote was a lot easier. But now it is like in 5 people at least 4
must be inclined through get you inclination vote and sometime even if
the 4 pimper people are inclined is still won't incline you because in some
teams there are like 2 to 3 interns interning in the same team.

Speaker 4 37:45
Hmm.

Speaker 5 joined the meeting

Speaker 4 37:53
Yes.

Speaker 3 38:09
So in that case, sometimes you even need 5 out of 5 votes for getting an
inclination.
But before it was like easier that if you even got 3 out of 5 votes, you will
be inclined and you will receive the offer letter as soon as you end the
internship. But now it it's not the same. The inclination criteria stuff, we're
getting an incline vote is tougher. And even if you got the inclination vote,
you might not know that whether you are gonna get an offer letter or not.

Speaker 4 38:14
Settings.
Yeah, that's true because you have a mental system in Amazon and a bar
riser system both. So your inclination depends on your manners manager,
mentor and then final decision is also approved by manager, by Barlaser.
And then there is a very senior RA or whatever is relevant to your role.
Who sits there, who is the head of the entire training thing. They approve
it once, only they all approve it 5A then only you get an inclination. If
anyone of them says no.
Especially the highest rates to the bar raiser at that time, if the bar raiser
seasonal, then you're off. You will be declined immediately.

Speaker 3 39:16
Nam, there was asking. This location depends on the position. No, the you
if even if you give the preference for anyone location, it doesn't. They
don't take your opinion. They'll send you wherever they want you.

Speaker 4 39:29
They just shake a die and tell.

Speaker 3 39:31
Yeah, exactly.
There were people who had put Bangalore as their like first first option
and there are people who had put Hyderabad as their first option but still
they got Bangalore and there were people who were put Bangalore as
their first options. They still got Hyderabad. So it's completely random.

Speaker 4 39:50
It depends totally on what the vacancy is, which team will is going to hire
you does not necessarily mean people who interviewed you will be hiring
you. There is a very, very big teams in Amazon, like most of us, into were
interning in the RBS team itself. But somewhere in Chennai, somewhere in
Bangalore and somewhere working remotely, I was also in RBS. I was
working remotely. Rohit and Nikita were in Bangalore working for RBS. So
was Sakshi there then.

Speaker 3 39:58
Yeah, true.

Speaker 4 40:20
And there was also working in the same team, but she was in Chennai, so
teams are quite big there. It's divided over daughter's locations etcetera.
But her team, Chennai, Hyderabad, Bangalore.

Speaker 3 40:39
And if you are like applying for roles, you must take tips from check for
getting an inclination would because in array there have been maximum
people who have been declined in bias. I think so it is relatively easier.

Speaker 4 40:51
Ohh.
So yeah, that's it.
Because in total they only hire 22 hours from the entire APAC and out of
which only 8 were inclined and out of 8 only 3 were non engineers.

Speaker 3 40:54
In.

Speaker 34 left the meeting

Speaker 3 40:58
Hmm.
Okay.
With.

Speaker 4 41:06
Who were not from VIP guest were also from Vt only VIT and engineers got
thin client and then there was Sandhya me and one goal or guy from
symbiosis who were inclined.

Speaker 3 41:07
Yeah.
So if you are really confident that you can crack it, then only I think so.
You should opt for array.
Because the intonation vote getting inflation vote, there is a lot tougher
and for.

Speaker 4 41:31
We can't open. I don't go to Tesco.

Speaker 3 41:33
Yeah.

Speaker 4 41:33
Quite a lot of.

Speaker 3 41:34
Just about a lot of Harriet and plus up talking about what happens after
decline, so if you're declined, it ends there itself. You won't receive a PR
anything from Amazon, but if you're inclined, you might receive an offer
within one year, and if not, then you have to again apply to Amazon
externally. Like they don't give you guarantee. But if you're in client error,
they relatively high chances of getting a PPO. But if you're declined then
there is the scenario of PPO and.
That itself.

Speaker 27 left the meeting

Speaker 4 42:13
All in all, given the current situation, I understand your spectres key hoga
by job security near the internship is one of the best internships that you
can get on campus in terms of what you can learn, how much you can get
to meet new people who are way, way at a different place than you when
it comes to experience, and especially now that a I is advancing so much,
Amazon is working very much on large language models, etcetera. If
you're very interested in that field, there is no better place than this.
But yeah, it comes with its own set of risk of being like numbers to them.
But yeah, the internship itself is very nice. All of us had a great experience
in turning their in terms of how the onboarding experience is, look like
entertainment and Biswa Kalyan.
Opponents.
Speaker 3 43:05
Yeah, true.

Speaker 4 43:05
Bought us, yes, but in that respect.

Speaker 3 43:07
That he also comes from rosters.

Speaker 4 43:09
Yes, that also happens. There's a good we have a good time at that places
too. But yeah, the work is also the work also takes a toll on you. It's a lot
of work there and complicated hard as well. But yeah, you will get to learn
a lot lot there.

Speaker 3 43:10
The.
The.

Speaker 4 43:42
Great system of keeping regular meetings to get updated on your
manager will keep regular meetings to be updated on your progress and
you can also ask them to do the same. You're mental will also be much
interested and then you don't have to run after the person that. Please
explain me this or that even in virtual this thing they are very very good
with that.

Speaker 3 44:06
So guys, I'll ask now rutik to go ahead. He is our super senior in Amazon.
So you have to think you can go ahead.

Speaker 5 44:17
Hey, thanks.
Ohh so free for about me. I'm also from I'm. I mean I'm your super senior.
A batch must be knowing me. But first starts. So I joined Amazon as a 15
last year. Previous to that I also joined as business analyst intern itself and
upon securing Pi joined US research analyst so I guess.
No, I thought well.
Uh must have explained you. What is the difference between B&R rules?
So just a few things that they might not be aware about. I would like to
add.

Speaker 12 joined the meeting

Speaker 5 44:57
Ohh yeah, so as said that it like most of the engineers got selected
because The thing is now this R role is converted as or in in a tech family,
right? So they expect that you must be having some kind of engineering
exposure or a I mean they would be expecting you that you have
inclination towards take part because it's a take job family now so.

Speaker 35 joined the meeting

Speaker 5 45:25
Ohh, I would say instead of thinking of if you are sure about going for take
or not. Rather you think if you are interested doing take work. For
example if you really enjoy doing a.
Engineering kind of stuff like if you you have 40 on coding or if you enjoy
doing. I mean it is not just like fitting some models by using whatever
existing in the system right? It is you you must be knowing or you must be
aware about how things are actually written from scratch. You must have
that interest towards a coding and stuff. So if you feel so then you can
look for a R rule. Yeah. And it is like if you are.
Naughty. Then there are high chances that you will be working on email or
data sense part and BS like some of the BS might work on ML projects or
some of them might work on automation or core analytics.
Of both the roles are quite great in exposure.
Ohh, you can definitely explore the opportunities within both. You have
ample of opportunities are to learn through internal courses and like it is
nothing like if you are BA then you cannot learn something from ML or
something or if you are R then you cannot learn for analytics part so that
won't be their learning would be happening in both the.
Dreams are both the rules, but later on again I'm saying it is just about
what you are inclined towards. If you enjoy doing analytics, then I think Bo
would be more apt for you. And if you are more into the text act.
Ohh or if you have more looking for that stuff then you can look for our
error.
Hmm.
Apart from that, I mean I I have completed an year now, so it was a good
experience for me. So being a part of such use organization, you get to
learn how things are really planned within an organization. How?
The system changes so quickly and you have to adapt it again. It is not
like every time you will be working on something new from scratch, right?
Somebody might already have worked on it since the scale is huge and
you just have to expand things.
Ohh for it or you just have to scale up the things so that will also be there
so.
Ohh yeah, I mean all sort of experiences you would get if you join and.
Yeah. And one more thing I would like to add. So with your technical
knowledge regarding the projects, you will be doing a one of the main or
one of the important aspect is if you're good fit for company or not based
on the LP they must be must have spoken about the leadership principles
that we follow and we do follow it on everyday basis. We do take decisions
based on both.

Speaker 19 left the meeting

Speaker 3 left the meeting

Speaker 5 48:27
Like even before, I mean all rather progresses quantified based on this LP
majorly. So it is very important aspect while you're evaluation of post your
internship to secure if you're there. So you should have those like it. It is
not like if you have you have done a great work but if they feel like ohh.

Speaker 18 left the meeting

Speaker 8 left the meeting

Speaker 5 48:54
Your way of work is not.
In accordance with the LP is that we have then they might not give you an
implant room. So yeah, so that is also an important aspect like you should
keep in mind while working on this like you have to explain whatever work
you are doing in terms of those LP. So yeah and the same thing would be
happening for your internship interviews as well. So one of the round
would be totally dedicated for that. So there also maybe you can create a
one scenario for.
You tell me and you can explain it based on that. I I joined sometime back,
but I'm sure that these folks must have explained what these LP I mean,
what row, what? What exactly they asked in interviews regarding this bar
is around or regarding leadership principles. So yeah. So go through the
Amazon website, you will get all the information about these Lp's and
what can you expect in that particular round so.

Speaker 36 left the meeting

Speaker 36 joined the meeting

Speaker 37 left the meeting

Speaker 5 49:57
Ohh do concentrate on that part as well. While you are preparing from
take perspective.

Speaker 35 left the meeting

Speaker 5 50:05
Ohh yeah, this is it. If you have any more questions this week please.

Speaker 6 50:29
Hello. Can I speak in night?
OK. Hello. Can you hear me?

Speaker 1 50:36
Ohh yes.
Yes you are.

Speaker 6 50:39
Yeah.
OK, so firstly, good evening to all.
Ohh my name is Parag. I was interning as a business analyst intern at
Amazon. So my experience with my experience of interview and
internship is totally different from like the like from a venture check. So
first thing first for the app.
Don't worry too much about it and don't overburden yourself.
Um, like whatever you can do, prepare it. Whatever you are. Strengths.
Focus on that first, because now, like, if you focus on other parts, then you
will get confused and you will also forget about your strengths. So focus
on that first.
Focus on SQL. SQL is your main thing.

Speaker 38 left the meeting

Speaker 6 51:34
To clear the interview for BA at least.

Speaker 39 left the meeting

Speaker 6 51:38
Ohh so my like that that's for apt then for machine learning part like there
was a part in aptitude so I must have shared you machine learning modes
back like they can send 2. So if you have those those were the sorrows,
ARS notes. So which we referred as a class those nuts. Those nodes are
like more than enough to cover like ML part. If you read if you have a time
then go and read about that.

Speaker 8 joined the meeting

Speaker 5 left the meeting

Speaker 6 52:10
And if you need those notes, then ping me. I'll give you the link. Second
thing about the interview.
So my in my interview my like the guy which was speaking my interview
was the Europe Head of analytics like Amazon office like from there. So he
haven't asked me anything from my resume. He my my entire interview
was totally like my first interview was one hour. So it was totally SQL.
And in like last 10 minutes he asked me one question about my like he
asked me like about like is there any rules which are like position of risk
responsibilities I have taken so I mentioned him about my Toastmasters
experience and my CR work. So he was curious about that. So he asked
me 23 questions like how you handled the situation and all but my entire
interview was based on SQL and I have been asked SQL in an SQL. I have
been asked.
Joins then. I've been asked common table expressions which is CT. Then
I've been asked Subquery and I have to write the codes like accordingly.
So I have written all the codes the codes in there is a time you will receive
a link. We have to download chime and your interview happened on that.
If like Amazon is following the same pattern to I have to write those
queries and like I have to share my screen my camera was open and I
have to write all of those things on the.
Time chat and I have to send it to him. So like I have been asked I think 7
question in first round out of those 75 were the coding SQL coding
questions and 2 were verbal questions. So he asked me like he first given
me a like he gave told me that there is a data you need to sell join that
data and then you need to outer join that data and how much rules will
going to be printed.
Like in the output that I have to tell verbally and there was another
question on normalization which I was not able to like. Tell him like I I I
haven't. I didn't know exactly what the normalization was so he was OK
with that. In my second interview.
I think it was a guestimate case study and again SQL was there and this
time they asked me about my extracurriculars and position of
responsibilities again. So I have been given a guestimates like find the
total number of petrol pumps in Mumbai. Then I've been given second
estimate was how many kites were there like flying during Sankranti.
Thank you. So I I've been given these 2 guestimates then I've been given
a case study. So even though in my resume like I I worked on NLP but I
haven't put like kept in my resume.
But he asked me that how will you improve page rank of a particular
product on any ecommerce website. So since I did that project in my
summer internship, so I I mentioned it like I knew some part basics of it.
So I mentioned it accordingly.
And then he asked me like again the SQL queries, but this time it was
purely theoretical queries and it was related to joints and like what is the
difference between inner join, outer join? What is cross join? What is sell
join like that. So there was just like that and then he asked me about tell
me about a time.
Ohh, when you helped your colleague or your friend in your.
No class or in your internship, so I I I prepared those kind of questions
already. So I had some.
Story back of my mind, and I also asked time like I asked him. Can you
give me 3040 seconds to think so they give you time to think about it. So
all the one more thing all of your behavioral questions shall be revolving
around the star method even for your resume like star method is
situation, task action and result. So you have 2 mention like you have to
give your situation like mention your.
Situation first. Then you have to.
Tell what kind of task you're going to do, and then what kind of action you
take then, and then result. What is the conclusion? So I've created those
kind of questions like for behavioral.
Ohh. Ohh, wait around. Sai. I went to Google and whatever questions I got
I prepared around like 50 questions like like that. But then there was no
need because you can go for like top 10 most to behavioral questions.
Which came in Amazon so you can do with like it it is suffice.
And then.
Half day also.
Ask me.
About my Toastmasters experience like they they were asking me since I
was a part of Toastmasters Club, so they were asking me how will you
implement your analytics in your postmasters like in your this like
position.
So I I told them so they're there has been a lot of data which is piled up in
the clubs and all. So I will take that data and I will visually visually create
some dashboards which will help us to pitch new Toastmaster meeting
new people.
How they will improve their communication skills and all. So I I I already
thought about that. So it was an easy question for me, but you need
whatever you are going to put in like whatever you're going to say or
going to put in your resume, you have to have a story about that. Don't
create like all the Landy stuff, just have like some points in your mind. So
you will sail through these interviews and one extra thing I did was I.
And you you kind of like you listen to Ruti. So I approached him before my
interviews and I asked him for a one mock interview. That mock interview
helped me a lot. So I so he gave me a lot of suggestions.
Ohh which I was lagging. So according to that I preferred so I so my
preparation was then structure so you can go for your seniors and you can
ask them for more mock interview it will help you a lot and that's it. And
for my experience so my experience was little bit different.

Speaker 7 left the meeting

Speaker 6 59:12
Um, it was virtual internship at first because my entire team was in US
and.
My I have been assigned only dashboards at first my manager told me
that for next 4 to 4 and a half month you need to do a dashboards. So
what I did was.
Ohh I whatever dashboards I have been assigned to I quickly like
completed all of those dashboards within 2 and a half months. So I I
worked on even on Saturdays Sundays because for my semester food I
needed to show something because only dashboards we cannot commit.
So that's why.

Speaker 12 left the meeting

Speaker 6 59:54
I I like when I completed all of those dashboards, so he was impressed. So
he told me what work you want to do next. So I told him I need like I
wanted to work on either machine learning or coding kind of thing. So he
gave me one extra project which was related to my dashboards. So I need
to fetch data from my all the dashboards which I've created and I need to
create a commentary.
Uh, like short text where you will get all the main like necessary points
which which will show you what is happening in those commentaries.
Because stakeholders don't have to go then in each of the dashboards
and see what is happening there. So like this was my project and like I I I
did this project and.
Um like before, 2 weeks like before the end of my internship. I.
I I didn't know whether I was able to do this project or not, but if you ask
your seniors and stakeholders where with whom you are doing your
project or anyone like which you know whom you know, so it will help you
a lot.
So that's why check was telling you that in the interview they test you
whether you will solve your problems or not, whether you are you have
that do not give up attitude or not because there will be a time that in the
internship you will get in like you will get stuck in your project and you
need to ask for the help. So that's it. So that was my overall experience. If
you have any questions you can ask.

Speaker 4 1:01:40
Also one more thing, I just like to ask they.
If you have questions from them too, please ask them during the interview
as well.
Because, like Barack, also mentioned that he only got the project because
he asked for it. He completed his work. Something very similar happened
to me in my first project was ended in 3 weeks because the data was not
significant enough and I asked. I told my mentor very clearly I have to
present a thesis to my college for my master's graduation and so this is
ending. I cannot work on this. I cannot show them this. So of course they
will put you on some minor projects if they something.
So we ended up quickly, but unless you ask them.

Speaker 14 joined the meeting

Speaker 4 1:02:23
They won't push themselves to give you another project or something like
this, so it's the same as Hogwarts. Help will be given to you when you ask
for it.

Speaker 6 1:02:39
One last thing I wanted to mention about the inclination part so.
You might be lucky, like for. In my case my bar razor, my skip level
manager and my manager were the same so I like. I never prepared 20
pages of documents which like these 2 guys have prepared. So I just have
to prepare only 3 page like 2 and a half pages of document.
Even like that also has not seized by my manager. So because I've I was
in touch in my manager on a weekly basis and this is the tip, once you get
in the Amazon, you have to have a communication with manager at least
once a week or twice a week. So if you are updating them regularly they
will know your process exactly like what, how how much work you have
done. So you don't have to like present it at the end.
And if your manager and Stiffler manager and bar is are same then you
you like you don't have to do and like nothing you have to do. You have to
just do your work and like in the last week like I was just chilling and that's
it and.
Ohh also do keep like after completing first 15 days you can keep one
note where you can like mention what are the important things you have
done, what are the like, what are the.
Uh.
Do you mention you have done in your internships so it will be like in
when you have to create a document of 20 to 25 pages you can highlight
those because you will forget everything after like whatever you have
done in first 2 months or 3 months. So you need to collect all these data
points so such that you can use it when you have to create a document.
OK.
And then have any questions?

Speaker 4 left the meeting

Speaker 1 1:05:15
Guys are there anymore questions?

Speaker 40 left the meeting

Speaker 1 1:05:22
If there's not, so, just find out the session.
Was still like to.
Send my huge gratitude on behalf of both the courses to our island, Nice
for their time today. They have definitely share their knowledge about
every single aspect starting from aptitude till the interviews, right up to
working there and I feel it definitely has helped us a lot today and also big
thanks to the placement team of other courses for organising this session
and also the audience for being so.

Speaker 17 left the meeting

Speaker 41 left the meeting

Speaker 31 left the meeting

Speaker 1 1:05:58
Easy.
Ohh the best friend Mazon and have a good day.

Speaker 22 left the meeting

Speaker 42 stopped transcription