Behavioral interview

Describe a time you had to learn something new to complete a

task.

S:duringasummerinternshipIwasassignedaprojectinvolvingRadioFrequency.
T:Thetaskwasformetocreateasoftwarethatreadsradiofrequencystrengthfroma
remotecontrol.
A:IresearchedsomeexamplesontheInternetandpracticethemuntilIwasconfortable
R:Asaresult,IwasconfidentenoughIcompletedtheprojectandfinisheditontime.

What is one current innovation in technology that excites you?

OneinnovativethatexcitesmeisSparkbecausethroughsparkwecanmanipulate
billionsandbillionsofdatawithaslittlecode.

A time you avoided an obstacle

S:Iwasworkingonateamprojectwiththreeotherstudents
T:Onestudentstoppedbeingproductive
A:Iproposetheteamtorestructuretherepartitionoftheprojectsowecouldmoveahead
R:Asaresultwewereabletofinishtotheprojectontime

A time you had to request extra help to complete a

project/task

S:IwasworkingonaschoolprojectandIwasstuckonsomemultithreadinguser
interfaceissues
T:Irequestedhelpfrommypatternwhohadmoreexperiencewithit
A:Ilearnedenoughfromhimandbenefitedfromhishelp
R:Asaresult,Iwasabletofinishtheuserinterfacewithinthedeadline

Describe a time when you had to implement a new technology

in your workplace/studies.

S: During one of my internship I had to design a storage database for

recruiting
T: I worked closely with the HR to gather the requirements
A: I implemented a SQL storage system to allow data attraction from
resumes
R: As a result I finished the new system on time and she gave me great
feedback

When do you stop learning or have learned enough?

I know that I have learned when Im able to move forward on a project

without being stuck or looking things up.

What is your weakness?

My weakness is my attention to details. I have had teammates that tell

me that I spend a lot of time on gathering requirements and
specifications.

Tell us about a time you manipulated data

I worked as data analyst for a company called Kingland System. My

experience there involved writing queries for extracting data, analyze
them and make recommendations.

Name one or two technical concepts you learned and how you
applied them

I learned spark for the seek of being curious. I had a project where I
had dataset from tweeter. Using SQL was going to be too slow so I
used Spark to extract the data.

What experience do you have with extracting data? (SQL

queries)

I worked as data analyst for a company called Kingland System. My

experience there involved writing queries for extracting data out of the
stock market, analyze them and make recommendations.
What is big data?

What interests you about data?

What interests me about data is data the challenges that come along
with data analytics. Nothing is usually the same and that excites me.
What is your experience in using Excel ?

MyexperiencewithExcelinvolvesgraphingdata,drawingfunctionsfrominputdata,as
wellasmodelingdatainExcel.

Case interview

Simple interest, compound interest etc. For hacker Rank 2 SQL

questions and 2 python questions:

Simple interest = Principal + Principal(Rate*Time)

Compound Interest = Principal(1+Rate)^Time

-Over Time times (years/month)

Simple profit/loss questions based on the credit card company working

strategy: case study: credit card insurance profit per customer cost

of mailing $0.5, response rate 1%, average balance per month $1000, Insurance
charges 1% of average balance monthly, customer claim rate 5% (once insured
customer files a claim, the balance will be wiped and C1 lose $1000). Question:
1) Profitable? Profit per customer,
2) How make more profitable?
3) What if response rate doubled but claims doubled?
4) Make chart of profit curve
5) Graph claim rate vs response rate
6) Graph profit vs response rate and find the response rate for maximum
profit (r=1%)

How to determine APR with Credit cards, based on data given

Simple interest = Principal + Principal(Rate*Time)

Compound Interest = Principal(1+Rate)^Time

-Over Time times (years/month)

C1 customers drop out in the middle of using their mobile app.

how would you analyze and report?

I would analyze using funnel survival guide. Its a tool that helps us see
the process and steps taken by costumer by giving us a visual
representation of the dataflow between each step. This tool will help
us to identify exactly where the costumer dropped out. From there, we
can develop a report with recommendation.

What are some ways you would detect credit card fraud?

-By flagging large expenses

-By flagging large expenses that are made after small expenses
-By flagging purchases that do not match owners usual
-By flagging when the card is used in a different state or country
without the card owners call.

Originally asked: If the ratio of green to red is 2 to 7 and there

are 70 total marbles, how many green are there?
Amended question after email: If the ratio of green to red is 2
to 7 and there are 70 red marbles, how many green are there?

7 green 2 red
g? 70 red

g=70*7/2=245

Given the dataset you need to perform analysis and give

recommendations

What are you gonna do if there is only 100 data for developing
model about prediction of male or female?
It will depends on what I have available: such as First name, weight,
height.
Most likely I would go based on the world statistics prediction

Technical problems: CI, missing data, p-value...

CI: Costumer intelligence is information derived from customer data.
Missing data: is an important element of a dataset that is missing that
we need to conclude decision on a dataset
P-value: is the probability for a given statistical model that when the
null hypothesis is true the result will be the same as more extreme
then the actual results.

Doing a project about New York City taxis data. Firstly do some basic summary
statistics and box-plots. Next one is to ask you to identify which trips are from/ to
the three airports in NYC. Then asking you to build predictions models for tips to
the driver. There are like 20 variables include continuous and categorical
variables. Last question you have 4 options, I choose the data visualization
one, which is plot the pickup and drop-off locations for taxis, but also need to
customize some settings to make it more informative.

Case 1:
Given APR, Interchange fee, Avg monthly balance, Avg spend every month,
and loss rate of 3% calculate the profit per customer.
Now justify if it is profitable to give cash back to the customers..

Case 2:
2 ways of campaigning for credit cards
1. Email - 10% of applicants become customers - each representative can
verify 10 email applications in an hr and is paid $25/hr
2. Chat - 20% of applicants become customers - each representative can
respond to 4 applications in an hr and is paid $25/hr
Profit per customer in both the cases in $100.
which one is profitable email or chat.
Draw the graph of profit vs no of applicants
Consider a scenario where there are only 5 representatives to handle
applications.
In this case which one is more profitable email or chat.
calculate the breakeven point for the no of representatives where chat will
be profitable than email.

-An ATM makes $2 per customer for 30% of its customers. Each time a
customer uses the ATM it costs 2cents. The ATM servicing costs $3000.
What is the break even point for this ATM?

-Determine average daily cost or profit, then tweak the initial set of
variables for new scenarios and finally determine a break even point.

-1 case interview was based on basic profit equation. profits= revenues- costs.
case was based on credit card scenario. credit card user population is divided in
to two groups. for group 1 I was given all numbers and asked to calculate profit
per year. For the second part i was asked to calculate the how much a different
group has to spend to generate the same amount of profit. we need to make
some assumptions here. I made some mistakes and the interviewer help me thru
the case.

- 2. Second case was interview was based on a campaign and calculating the
benefits of campaign. Interviewer asked me how I would calculate the profits
related to campaign. he was expecting an answer like " I would have a target and
control group and based on the difference in the increase in revenue before and
after the campaign I can tell what % of the increase in revenue is due to the
campaign". something like that.

-Analysis of a shipping company's balance sheet

-Comparing credit card collection strategies

-Draw a line that would give domino's pizza the same revenue $9 per
pizza, with the y-axis (% On Time), and the x-axis ($ off for being late)
-An interesting Data challenge that can be completed with any open
source tool.

-A coding round that has SQL and basic programming. And 2 case
studies during on site round out of 4 total.

- sql joins and programming questions on hackerrank. Next was a data

challenge - you are given data and a week to build a dashboard.