A tool kit for searching your first

job in Data Science

A wrap up of things I learned from my own
experience

Data science is a very broad field. In the beginning, the data

scientist was a one in all positions. You probably had to know
engineering and machine learning. Be an outstanding data analyst
and fluent on the cloud. Write SQL and a couple more
programming languages as a developer. Or even better, be a
proper developer! It was overwhelming. So when the field started
to expand, much of these positions obtained their own
independent role.

These, my friends, isn’t something that happened gradually as it

worked out in other industries: it has changed dizzily in a matter
of perhaps a couple of years. The field has found that what we used
to call a lot whole of data, can be now found in a smartwatch and
there’s an actual need of bigger teams and more dedicated
professionals for all the different kind of tasks.

Mind that when we talk about the expansion of the field, we’re not
referring to a merge of two disciplines like Business Intelligence
and Statistics. A lot has been talked about this, and even though
it’s used to say that Data Science borns from the fusion of domain
knowledge, statistics and business intelligence, I personally think
different and probably the best graphic interpretation I’ve seen
comes from the blog Upside:
The image is pretty clear by itself and shows how Business
Intelligence and Data Science are two completely different fields.
Meaning that the expansion of the latter comes from the
individualization of tasks under different roles, instead of from the
transformation of Busines Intelligence positions into Data Science
roles. Nowadays you can find around the same number of job
posts for roles like Data Engineering, Data Scientists or Data
Analyst. Having also afterwards other roles in full expansion, such
as Machine Learning Engineering or Data Translator.

All this has brought people from different fields -like me, into Data
Science tempted by the exponential demand for more and more
professionals, but also because of the opportunity of getting
involved in a relatively new field. And a pretty much exciting one if
you’ve always liked data and working with numbers, and you enjoy
trying to understand why things happen.
If you feel identified with the description above, and you’re
thinking about pivoting into the Data Science industry, or perhaps
you’re already halfway through, in the next tips I’ll try to share
with you some stuff I learnt from my own experience when such
not a long time ago I decided to fully get into this path. Hopefully,
it will be useful for you too!

1. Know yourself
As the field grew up, the diversification of roles made it easier for
people to get into the industry under some of all the roles
available. So the first important thing to do is to understand for
which of all the possible roles am I right now ready for and in
which role will I like to be in some years.

The three more common roles found right now online are
probably:

 Data Scientist: the guy usually in charge of building

the machine learning models, often including most of
the features to be used by it. They in charge of constantly
improving the performance of the model, and trying new
ones in order to improve the predictions.
 Data Engineer: this guy is probably more technical in
programming and computer science. They are in charge
of all the data flow in order to get data from and to the
models built by the Data Scientist.
 Data Analyst: the Data Analyst job is probably the one
most misunderstood and confused throughout the field.
When I was looking for a job, constantly found job posts
looking for a Data Analyst, requiring to know and use
only Excel, with no statistics or maths knowledge
 needed whatsoever, asked to find, describe and visualize
insights to the commercial or management team. Ideally
automatizing the process. These roles my friend, to my
understanding, are Business Intelligence position. The
Data Analyst, again, under my vision, is someone
constantly performing ad-hoc analysis, with a solid
understanding of concepts such as distributions,
probability or algebra, regularly exploring patterns in
order to improve the machine learning models and the
whole company understanding of how they work. And
sometimes working out for the Data Engineering which
data will be needed and how to get it.

If someone felt that some of the descriptions above were not

accurate being a Data Scientist, Engineer or Analyst, please don’t

take it on me 🙃. As I’ve said, the field has grown almost too quickly
to follow, so probably there’re combinations of some of all those,
under the same or different role names.

If you don’t have clear right now which one should be your current
and future role, don’t worry, it took me a long time to understand
mines. For example, I call myself a Data Analyst who can do
Machine Learning. I enjoy both things, but I know I have more to
offer right now to a Data Analyst role. Will I move into full-time
Data Scientist position in the future? Who knows. On the other
hand, you might be someone with a strong computer science and
programming background, capable of starting as Data Engineer,
but knowing you’d like to eventually move into Data Scientist. To
my understanding, the possibility of learning and pivoting within
the field is very high.
Source: Unsplash, Elena Koycheva — Free stock images

Practical tips:

 Work out a personal brand statement. This might

sound silly, but it can be very helpful for understanding
yourself to put down in words, in a short paragraph of
maybe 4 or 5 lines, your career journey, your main skills,
your achievements, goals and why would you be
valuable for a company. And yes, I did say 4 or maybe 5
lines. I know it’s very little, but that constrain helps you
to go only for the most important stuff.
 Write down your professional story: as it is big the
demand for professionals, it is also big the offer.
Meaning you’ll be competing with some very passionate
lads. So if you’re someone like me, or perhaps you’re
coming from a different industry or mixed experiences,
be sure to be able to explain your professional story, as
well as how and why did you arrive into Data Science.
2. Getting ready for the job search:
the most important tools
Previously to start looking for a job, and while I was doing it, I
heard of several websites and stuff I needed to have in mind in
order to succeed. However, in practice, I found very little of these
tools were really needed and useful for me to finding a job. And
even most important, in a daily routine of at least 6hs fully
dedicated to the job search, I found that having a constrained
number of things in mind was very important in order to be able
to cope with everything.

God save the king LinkedIn: it is not only the best place to
look for a job, according to the number of job posts and the
activity level of the people looking for employees, but it is also a
great place for networking and show yourself. The main tips I
would give to someone just getting started are:

 Build your profile: nice photo, complete professional

experience and education, and good spelling/grammar
are the basics. From there you have stuff such as
customising your URL -instead of using the default URL
LinkedIn gives you, selecting a proper background
picture, writing down a nice headline, showing up your
updated contact information, including a good and solid
summary (could come from your brand statement) and
filling as much as possible all the other sections included
by LinkedIn such as skills, accomplishments or
recommendations.
 Keep it up: try to maintain active within the network,
showing yourself to your contacts with Data Science
related content, but also constantly adding and
 following up people in the field to increase your network
of contacts. They might be useful if they post any
opportunity or maybe in case of finding a job post in
some of the companies where they work.
 Tune your settings: when your profile is 100% complete,
it will be time to switch on the option “Let recruiters
know you’re open”, as well as using your free trial of
Premium. Wha by the way, personally, it’s something I
fully recommend since gives you the chance of searching
and stalking more. Probably the two things you will be
doing in LinkedIn the most during the following weeks.
 Reach to the right audience: connect with people in your
city, ideally, in the specific industries you’re interested
within the field. Don’t loose time adding CTOs or CEOs
from fancy organizations. These people won’t be even
close to the hiring process you’ll be interested in. Instead
add Senior Data Scientists, Leads or even people from
HR, who might be involved in processes more according
to your profile. Last but not least -and this is
probably the best advice I received when I was
looking for a job, add alumni from your school
currently working in the field. They’ve already gone
through all this and lots of them will be glad to help you
with recommendations or at least a good piece of advice.

This is my LinkedIn profile in case you’d like to take a look at it.

Very mind your Github: GitHub is not only a place to have all
your shared and personal projects online, but it also shows stuff
about you. It shows how active you are, how you code, the kind of
things you like to work on and a basic description of yourself. So
upload a photo, include some basic information about you, and
most importantly, be sure to push a wide variety of your projects,
showing a different kind of skills and interests. Also, ideally, push
to GitHub as frequently as you can. Part of looking for your first
job in Data Science is like learning a sport: you can’t stop doing it
for a while, otherwise, you will forget stuff. So be sure to be
constantly pushing your daily work, in order yo show yourself
active and improving your set of skills. This, for example, is my
profile.

An example of my own activity panel on GitHub.

Your CV does matter: the curriculum vitae is still pretty much

import nowadays, and even though I won’t give advice about style
and design since it can be very subjective, I will give you some of
the best tips I’ve heard:

 Keep it short. Ideally 1 page, at most 2. Even if you’ve 8

previous work experiences and a PhD. People just don’t
read more than that.
 Silly but important: include all your contact
information, and include it correctly. Double-check your
phone, email and all URLs.
 A personal bio at the beginning is important. HR people
and hiring managers receive hundreds of CVs for each
job post. Give them a quick reason to keep reading your
whole curriculum.
 Include your skills. Instead of making someone decipher
your whole set of skills from previous jobs descriptions
 and education, make it easier for them and be explicit
about it. Also, some companies use software to identify
keywords through curriculums and filter them. So make
sure the important stuff is mentioned at least once in
your cv.
 Details matter. Good spelling, text alignment,
scannability, name of the file you’re sending…all this
stuff talk about yourself. So be neater than ever.

3. Where to look
The mission here will be to be as efficient with our time as
possible. Lots of the job posts websites around ask for you to fill a
very very long ad-hoc profile before even starting to search and
apply to their job posts. Also, much of the companies post their
offers in several places, so a lot of the job posts you’ll see around
are going to be repeated in between sites. So be wise with your
time, and instead of spending hours looking for job posts through
several job boards, choose a couple and focus on them. For me, the
main job board as mentioned before is LinkedIn. The second one
might depend according to where you live or the kind of
companies you’re interested in. In the UK Glassdoor or Indeed are
both good options. But if you’re looking to join a Startup, Angel
would probably be your place.

The main point though is: don’t lose your time. Daily you
should be less than 50% of your exclusive job searching time
looking for the right job posts, and the rest of your time applying
to them, including filling up your data on lots of companies
profiles and writing nice and pretty cover letters. That can sound
like a whole lot of time but remember you should also be working
on side projects, maintaining yourself active on LinkedIn
somehow, preparing interviews eventually and also having a life,
so you can see why we don’t want to lose any time if not worth job
boards.

4 Application strategies and keeping

track of all
Looking full-time for a job can be more stressful than people
generally think. That’s why it will be important for you to keep
organized:

 Start by breaking down your week. Monday and Fridays

are usually bad days for applying or reaching out to
people. So take advantage of those days for working on
your side projects, writing cover letters, studying
something or looking for people around.
 When I was looking for a job people at school told us to
aim to send in between 10 and 20 applications per week.
Don’t know where the number came from, however in
practice, it felt like an aggressive and exigent goal, but
completely doable and productive.
 Keep track of everything. Did you send an application?
Added someone to LinkedIn? Wrote an email to a Data
Scientist you met in a MeetUp? Whatever is the thing,
find a tool o method you’re comfortable with, and write
it down. In my case, the tool was a simple Google Sheet.
Each time I was reaching out to someone or applying for
a job post, I was writing down who I contacted, how I
did it, the date it happened and some more useful data
for myself. I was doing all this in order to be aware of
what I had done, but especially, to follow up each
action after exactly one week if I had no notice
 by then. Don’t be afraid of doing this, people usually
don’t take it badly at all, since anyone can busy as to
miss an email or a private message.
 Go that extra mile. Whether you’re applying to a job post
through a job board or a company website, be sure to
write cover later (short and sweet: two paragraphs…one
about you and one about the opportunity and company),
and ideally sent it directly to someone in the company
through email or LinkedIn. And you don’t have
Premium in the latter, don’t worry, a short message with
the right words will be better than nothing at all.

A lot has been said about the cover letters but in my experience,
even if no one reads it, it shows anyway something about you if
you took the time to thoroughly write something about the
opportunity you’re applying to.

5. Prepare your interviews

Not going prepared for an interview is almost like not going at all.
In my opinion, these are the main things you always need to
prepare:

1. Technical questions: there’s a set of basic questions

in the Data Science world that are asked especially in
first interviews. Won’t go through this list on this story,
but you can easily find on Google several sites listing all
these questions.
2. Competency questions: this is more about you…
previous experience, education, skills, strengths,
weaknesses, motivations, etc. Again, lots of sites online
 have examples of these questions for you to prepare
yourself and give the best version of you in case of being
asked any of them.
3. The company and the role: employers want people
motivated and interested in the company, so be sure to
research it as much as you can and try to demonstrate
that knowledge throughout the interview

As mentioned before, all these points and suggestions try to be a

wrap up of all the stuff I found useful during my own job search,
but also from looking at the job search of several of my classmates.
However, I’m sure that lots of other people might have had a
different experience with some of the things I’ve mentioned on
this story, or maybe they have something extra to add to this job
search tool kit I’ve tried to compile here. If that’s the case, please
be sure to share your experience on the comments, so we can all
learn from that too!

And remember, Data Science is a field constantly changing. It has

changed a lot in the last year and you can be sure it will keep
changing in the years to come. So keep moving! Listen to podcasts,
go to meetups, read as many books as you can, keep always
studying new stuff, and of course, stay tunned here in Medium…
it’s an amazing source to keep yourself informed.

If you enjoy this story don’t forget to check out some of my last
articles, like how to divide your data into train and test set
assuring representativeness, survivorship bias in Data
Science and an introduction to NLP and 5 tips for raising your
game. All of them and more available within my Medium profile.
And if you want to receive my latest articles directly on
your email, just subscribe to my newsletter :)