08 Jun Data Science Career Advice from Industry Experts – Part I
As the demand for analytics talent continues to explode, career opportunities for data scientists seem endless. We see that for people that are just starting out on the journey, there are many questions to be answered and it always helps to seek advice from experienced professionals.
So we gathered a small sample of ‘words of wisdom’ from Quora, KDnuggets, and Medium, experts sharing their opinions on data science careers. Hope this compilation sparks more interest for people who are either curious about what’s so sexy about data science or want to be better at what they do.
Finding the Right Match
One of the important things to do in this sea of possibilities -other than loving “data” itself-, is understanding the differences between data science roles, finding the ones that excite you more to explore further.
“Figure out what kind of data science professional you want to be: This may sound a bit preachy, but it’s very important. To achieve your goal, you need to have a solid understanding of what your goal actually is! Traditional analyst, data infrastructure, predictive modeling (machine learning), or deep learning & AI. All of these require different skills, so you should think about it early to pick up the right skills.”
-Joyce Wu, Data Scientist at Quora
Vik Paruchuri tells his story to highlight the importance finding an area that you are passionate about.
“My entry point to data science was predicting the stock market, although I didn’t know it at the time. Some of the first programs I coded to predict the stock market involved almost no statistics. But I knew they weren’t performing well, so I worked day and night to make them better.
I was obsessed with improving the performance of my programs. I was obsessed with the stock market. I was learning to love data. And because I was learning to love data, I was motivated to learn anything I needed to make my programs better.
Not everyone is obsessed with predicting the stock market, I know. But it’s really important to find that thing that makes you want to learn.”
-Vik Paruchuri, Founder at Dataquest
Back to Fundamentals
The theoretical framework for data science involves a heavy dose of mathematics and statistics and you will see many experts believe having strong fundamentals is crucial for many of the data-driven careers.
“Learn math, and look for the main point of a project, rather than a document of requirements handed to you. Both skills will serve you well in applying the correct algorithm for the data and the problem at hand. Both are also subtle, and it’s possible to gain employment and remain employed for a bit without them. However, both of these will help you go from an okay data scientist to a good or great data scientist, which is crucial for career advancement and remaining employed when standards are set for the field”
Colleen Farrelly, Data Scientist at Kaplan
“Spend (significantly) more efforts on the very basics: mathematics, statistics, linear models, and etc.
A solid understanding of the basics not only makes learning complex, advanced techniques easy, but also helps you understand better the strengths and limitations of those techniques.”
Boxun Zhang, Data Science Lead at GoEuro
And according to Yisong Yue, Math is just another language, such as a Basketball jargon or as the notes in Music!
“When I first started studying machine learning, I was overwhelmed by all the mathematical notation and abstract concepts. But over time, I gained a comfort level that allowed me to use math to reason very efficiently about different aspects of machine learning. After all, math is just a language, albeit one that is impeccably precise and rigorous. Biologists have their language, basketball coaches & players have their language, and musicians have their language. Language & vocabulary allows one to modularize complex concepts so that one can communicate more precisely & efficiently. That’s all that math is”
Yisong Yue, Machine Learning Professor at Caltech
Finding Your Unique Way to Learn
Learning is not a one-size-fits-all phenomenon, everyone has their own way. You can go by the book and find formal classroom training to be more effective for yourself or dive right into it by using the endless online platforms and data sets.
“Don’t obsess over theoretical victories, find real-world problems to solve: The world is full of problems — created by data (revenge porn — yes, there is research going on to tackle this, teenage depression, digital disenfranchisement) to unsolved problems (medical and mental diseases etc.)”
-Tarry Singh, CEO, founder & AI Researcher at deepkapha.ai
Monica Rogati has a very practical perspective that she thinks is a more efficient way to learn and according to Claudia Perlich, learning data science is just like learning ski: you need to learn in the process. Things may go wrong during that process but failure is just part of the deal.
“Many of these resources follow a common pattern: 1) here are the skills you need and 2) here is where you learn each of these. This makes sense; our educational institutions trained us to think that’s how you learn things. It might eventually work, too — but it’s an unnecessarily inefficient process. Instead, I recommend building up a public portfolio of simple, but interesting projects. You will learn everything you need in the process, perhaps even using all the resources above.”
-Monica Rogati, Data Science and AI Advisor
“I think, ultimately, learning how to do data science is like learning to ski. You have to do it. You can only listen to so many videos and watch it happen. At the end of the day, you have to get on your damn skis and go down that hill. You will crash a few times on the way and that is fine. That is the learning experience you need. So; volunteer your time, get your hands dirty in any which way you can think, and if you have a chance to do internships — perfect. Otherwise, there are many opportunities where you can just get started. So just do it.”
-Claudia Perlich, Senior Data Scientist at Two Sigma
“In this hyped field, there are many pieces of chicken-soup-for-the-soul advice (stay curious, be persistent, etc). As a practitioner, I find reading academic papers and technical blog posts (stuff like Neural networks and deep learning, Kaggle winners’ blog posts, Distill) much more valuable to my growth. Or more generally, content not filled with platitudes but with concrete examples and takeaways.”
-Lili Jiang, Head of Data Science at Quora
This article is originally published on IADSS website (www.iadss.org).