Breaking into Data Science: The Recap
Recapping Many Stories and Tips to break into Data Science
Recapping Many Stories and Tips to break into Data Science
Breaking into Data Science is hard. From this survey, most Data Scientist listed the core skills they have are Python/R/Excel, Data Visualization, Critical Thinking, Communication Skills, Machine Learning, Statistics, Math, SQL, Bussiness Understanding, and Data Preparation. Just look at those varieties of the skills to possess, it could be scary to learn all of them. Moreover, Data Scientist job advertisements posted in the online job boards often require a lot of experience and/or knowledge of the technologies that only accessible if you previously have been in the field for some time.
As hard it is to break into Data Science, it is not an impossible feat. Many people did in fact, and so do In this article I want to write the recap from all the stories I have read and listen from many sources on how to break in into the Data Science field and what tips these people give.
Breaker Categories
What is Breaker? It is just a word that I come up with for convenient purposes. In this article, I would refer the people who want to break in as a “Breaker”. While many people are Breaker, they could actually be categorized. There is one Reddit post that classified the people who want to break into the Data Science field which I quoting but I add my own views here. The classifications are:
1. Researchers
Breaker in this classification coming from academia or used to work intensively in one. Usually hold a Ph.D. (or Master in some countries); often from the quantitative field, have a track record for publications and conferences, and work as a researcher/assistant/professor/lecturer.
Researcher Breakers are great in the theoretical, but commonly have a hard time transitioning from academia into industry. Most of the time they also lack the business perspective and burned out by how different the responsibility is.
2. Career Changers
Career Changers Breaker is a Breaker who already has sufficient experience as a professional in the Industrial, might hold a master's degree or not, and willing to change their career to be in the Data Science.
Depend on their previous experience, this Breaker could have almost all of the core skills necessary to break into Data Science (Often is Programming, Communication, and/or Bussiness Understanding) to does not have any of the core skills at all. The only thing for certain is their lack of seniority in the Data Science field.
3. New Grads/Current Students
Just like the name states, this Breaker is a Fresh Graduate from College or currently a student in one. It could be in any bachelor major, but mostly this breaker coming from a quantitative or technical field.
This is the Breaker who often have the hardest time because they need to find that sliver Entry-Level position which overwhelmed by massive applicants. Moreover, many of this “Entry-Level” position is looks like not an Entry-Level at all, like they required some advanced technology or some similar experience. Although, I would argue the companies do not know much about Data Science if there is an Entry-Level position that seems to have many advanced requirements.
Story Recap
From the Breaker categories I just explain, depending on where you are standing, the obstacles in front of you would differ. Here I would recap how people breaking into Data Science depending on their Breaker category.
Researchers Breaker
This is me. I am coming from Academia background and I was a researcher before breaking into the Data Science field.
Most of this Breaker type have a common ground when they want to break into data science.
They love contributing to Science and want to work the theory for a better humanity. The problem is that Academia world is hard, and not as pure as it is. The environment for publishing and competing who get the name for those Breakthru in the science is put off many people from Academia world.
A rigid career path and only a select few peoples who could become successful. Moreover, financial safety and work-life balance would not compatible with people who consider building a family.
Many people have sufficient knowledge in the theoretical and/or programming algorithm but feel that this knowledge only stuck in a certain place. They want to apply their knowledge.
There are some obstacles that this Breaker face. Most of them are:
It is hard to transitioning from Academia to Industry as the environment really different. People from academia don’t have experience in an industry work environment. Working as a data scientist within a corporation requires an understanding of how the business world works. That means people who coming from Academia need to re-calibrate their minds from the academic culture to corporate culture. Different than fresh graduate that could be molded and experience professional, it takes an immense effort to change a mindset from academic to the industry.
People from Academia often hold a higher university degree, which means that there is a perception from the companies that they want to ask a higher salary compared to the undergraduate degree holder. Either it is true or not, this is the perception that they get.
Weak on-site programming skill interview and/or business communication skills. Many Academia people are the person who inherently loves writing and focuses on their research. The problem with Data Science is that, not only you need technical skills but you need to present it to the non-technical people on a daily basis. Yes, research in academia always including presentation but mainly it was presented into people with similar backgrounds and knowledge. This is the difference, Data Scientist would need to wrap the technological term into a more commercially selling term.
Now, how these people actually break into data science then? Academia people still hold an advantage to have an experience and theoretical base. Most of them said that what they need to do is convincing someone you can handle the job. This means that you have the prerequisite knowledge about Data Science and showing the experience and skill you have from doing similar work.
Knowledge could be acquired by a lot of reading, online courses, or having a mentor. Presenting the experience and skill would need a lot of adjustment; I personally tailor my experience in my research time to fit the business that I want to enter. For example, I am a biologist with experience in biological statistic methods. I just need to show my statistic method that actually appropriates in the industrial business in my CV and interview.
A lot of stories also pinpoint that they applying for the job via online job boards but not randomly apply. They apply specifically for the job with a research business background that they had (e.g. healthcare). Although, I also have read many stories (like me) who find employment from their own connection. This connection could come from their University, Meetup group, or just your old pal.
Career Changers
This is the type who already a Professional in Industry before but wants to change their career into Data Science for any kind of reason. This breaker commonly is people who:
They want a bigger salary. Mostly career changers breaker does not have the same motivation as the researcher breaker. Although financial security is still the target, the Researcher breaker also motivated by the way that their research work could be applied whereas the career changer breaker mostly just wants that bigger salary (not generalizing, just from what I read).
Want a better career opportunity. Related to financial security, Data Science is still a hot job right now with many demands in the industry. The career pathway is also clear enough which attracts many people to try break-in.
Betting everything in one bag. I personally had encountered many people who want to change their career mid-way especially those who attend the Bootcamp. The people here betting their last money and employment to have a chance to break into Data Science.
Professional often have better leverage as they already have the correct experience and mindset, but there are still some disadvantages:
Lack of Core Skills. Depends on their professional background, they might already possess almost every necessary core skill to none at all. For example, people who worked as Software Engineering would do better at the programming part but not necessarily when need to communicate their work.
They sometimes just do not have the analytic capability or the passion to do one. Yes, analysis. Don’t think that only Data Analyzer analyzes the data and Data Science only using Deep Learning to predict something. Data Science is heavily related to analyzing the data and do your creative work with it. I have read and met many people who were just not up for it. It is especially shown during the interview part where they need to showcase their analytical ability and just can’t.
Professionals with a few core skills would still have an advantage compared to the other Breaker, especially if they already have the domain knowledge and/or the programming skill. Many breakers from this categorize often showcase their professional experience during the interview and filling their lack of skill by a lot of reading and following online courses/certification. Building a Portfolio is also a must. Some also enrolling in a higher university degree (Often Master) to improve their chance. Others also recommend doing an Internship.
Just like Researcher Breaker, many people apply through online job boards but often time the career changer breaker more flexible in terms of the industry so they apply anytime there is an open position. I also read many stories that career changer breaker secure a Data Science job from their own companies. For example, I read their company wants to build a data science team, and because this person possesses the skill they hire them as a Data Scientist.
A lot of the career changer breakers also get a job as Data scientists because of their presence in online media; especially LinkedIn. They have tried blogging or publishing their work in the online presence and out-of-blue recruiter contacting them.
Networking although still the most ways that people recommend. With networking, they could bypass many of the usual steps if you just applying like you normally do and networking could also help you to soften your way into the position.
New Grads / Current Student Breaker
This breaker is might be the harder type to actually break in the Data Science field. Without an actual industry experience and trying to find a limited spot where the competition is fierce (not just from fellow new grads but also from the experienced). Most of this breaker has these following traits:
Getting a job in any Data related position is fine, but mainly aim for Data Analyst or Data Scientist position. This does not mean there is no specific aim, but because of the competition, lack of connection and the limited spots make this breaker fine to any kind position.
They might hold a bachelor/master's degree or currently pursuing one. Some not holding any degree at all, often fresh from the high school.
Have applied to many companies and even have an Internship before during student time or after graduation. Some also have called for phone interviews/face-to-face interviews/coding interviews or any kind of interview with varying degrees of results.
Some interested in Data Science because of the salary, other interested because of the technological prospect. There are also many other reasons but this mainly what I see.
Once more, I stated that this breaker has the hardest time to break into Data Science because:
Lack of Experience and Core Skills. This is always the main reason that this breaker has a hard time. Data Science is still a new freshly baked from oven occupation, means there is more demand compared to the supply for this position. This also means most companies and early-rise startups are looking for the experienced one.
No Proper Connection. For a fresher or student, connection to the industrial business is hard to find. Many of the stories I read is that people find their occupation through networking and often the comment by this breaker is “How Could I do Networking?”
There are not many Entry-Level Positions in the Data Science position because mostly they are prerequisite an immense amount of experience. Even more, I also read many of the fresher are having a hard time even just called for an interview to this supposedly called “Entry-Level” Position.
It seems that many hurdles to break into data science as fresher, and it is, but it is not impossible.
These fresher who manage to break in are those who persist in their effort and keep updating their knowledge by reading a lot and doing online courses. Some find employment from their time doing an internship, other by getting into the non-technical industry which has an opening for a data-related position.
Many also giving tips that if you don’t hold a grad degree it would be better to try finding an internship or try to pursue one. Some also suggest applying for another position but when you already had a sufficient experience you could try to break in once more as a Career Changer Breaker or Researcher Breaker.
Finding a connection as a fresher or student also not impossible, and in fact, many did. Many find one from their professor in the university, the job board from campus, or even just from their hobbyist group. People also try to connect via online media, such as LinkedIn. One piece of advice regarding networking is that it is still a human interaction with each other and it means to give and take. Don’t suddenly connect with someone in the meetup meeting or LinkedIn by asking if there is an open position or not. Try to have a proper discussion and being friends with them. Networking is not an instant way to secure a job after all. It needs to be slowly taken care of before any result could be produced. I personally build my networking since I was High School, and it only pays off almost a decade later. Networking, after all, is about trust with each other and to some degree, give and take.
Conclusion
Many people try to break into Data Science and depend on the experience they have, the obstacle would be different as well. A lot of tips by the successful breaker have been given, it includes:
Update your skill to match the data science core skills
Personalize your experience to match the position requirements
Doing an Internship
Networking