Advice from 15 Top Data Scientists
What are the essential skills to be a "good" or even "great" data scientist?
Well, I asked 15 data scientists from multiple companies and industries what they think, and let's say I was pretty surprised by some of the responses.
Hopefully, this article will give you better guidance on where to focus to progress in your Data Science career!
Results
To avoid wasting your time, here are the top 10 skills described by the 15 data scientists summarised in a plot.
What I find pretty interesting is the combination of technical skills like maths and coding and softer skills like communication and curiosity.
Personally, this makes sense. The true heart of a data scientist is to find trends and information in the data, and you need technical knowledge and tools to do this. Then, you need to convey that information easily to senior management and stakeholders so they can make informed decisions for the business.
It sounds so simple, but due to the volume and vastness of data, it is often challenging to know where to look. It literally is finding a needle in a haystack. Fortunately, there are tools and strategies to increase your luck, which we will now go over in reverse order.
SQL
SQL is the "language of data", so it makes sense why it is in the top 5. I don't think I have ever seen or heard of a data science, or even data, job where you don't use SQL.
In this "big data" world, Excel doesn't cut it when trying to analyse millions of rows of data, whereas SQL works seamlessly. Nowadays, pretty much every company has loads of data, so SQL is universally used by at least one person in each organisation.
Fortunately, for those of you at the start of your journey, SQL is relatively easy to learn compared to other programming languages like Python or C++.
This is because it's tiny and specifically designed to work with relational databases, so there is little to learn before you start to feel proficient and can start solving real world problems.
You are pretty much set if you know the basics, which are:
- *SELECT FROM** (standard query)
- ALTER, INSERT, CREATE (modify tables)
- GROUP BY, ORDER BY
- WHERE, AND, OR, BETWEEN, IN, HAVING (filter tables)
- AVG, COUNT, MIN, MAX, SUM (aggregate functions)
- FULL JOIN, LEFT JOIN, RIGHT JOIN, INNER JOIN, UNION
- CASE
- DATEADD, DATEDIFF, DATEPART (date and time functions)
I have a whole separate article explaining the exact SQL knowledge you need to be a data scientist, in case you are interested.
Python
No surprise here; Python is joint 4th on the list. Python and SQL are the power couple in the data science, Machine Learning, and AI worlds.
A lot of the ML ecosystem and libraries are designed with Python in mind, and like SQL, I haven't seen any data science job that doesn't require Python in one form or another.
However, unlike SQL, Python is a bit trickier to learn. It is a general-purpose language, so it can do many things and be applied to several areas in the tech space.
When it comes to Python, I recommend learning the basic syntax. These are things you should make sure you learn:
- Variables and Data Types
- Boolean and Comparison Operators
- Control Flow and Conditionals
- For and While Loops
- Functions
- Native Data Types (Lists, Dictionaries, Tuples etc.)
- Classes
- Packages
Then learn some data science specific libraries:
- NumPy – This library is designed for scientific computing, offering many mathematical functions and matrix support. Developed in C, it has optimized computations, which is particularly beneficial for handling large models and big data.
- Pandas – This is the go-to library for loading, manipulating, and working with data in Python. It is excellent for almost any data analysis task and is easy to use.
- Matplotlib – As a Data Scientist, you must visualize your data or results. Matplotlib is the leading visualization package in Python due to its wide range of abilities.
Aftet that, start working on projects immediately. Only through practice will you really understand Python.
I have a separate article detailing how I would learn Python again, in case you want to check it out.
Machine Learning
As a data scientist, you need to be familiar with ML. That's not to say you should be able to derive transformers mathematically from first principles.
However, you should, in my opinion, understand various algorithms across the supervised and unsupervised learning space and their advantages and disadvantages in certain situations and problems to which they are mainly applied.
You should also be familiar with concepts like overfitting, cross-validation, hyperparameter tuning and regularisation, all of which are best practices when building an ML model and are used all the time in industry.
ML is a vast field with constant developments, so it's unrealistic to expect to know everything constantly. However, fostering a natural curiosity, a top 10 skill, is invaluable in honing your data science abilities.
If you want to learn ML, the best course is Andrew Ng's Machine Learning Specialization on Coursera. This is the first ML course I took back in 2020. However, it has since been revamped and includes exciting recommendation systems and reinforcement learning topics. Not to mention, the coding tutorials are now in Python!
The next course I recommend is Andrew Ng's Deep Learning Specialization. This is a follow-on course from the Machine Learning Specialisation and will teach you everything you need to know about deep learning, which is one of the biggest reasons people often get into data science.
Check out my machine learning roadmap if you are interested.
Communication
The second most important skill for a data scientist is actually a soft one, but it's not to be underestimated. Communication is key in our field.
Without presenting and showing your work and insights to others. How else are they going to know about it?
Data scientists are the lynchpin between the business and tech side, as we typically know both domains quite well. Therefore, we can tailor our presentation style to the given audience and see things from both sides.
The problem is that communication is a challenging skill to master, which you will constantly improve forever.
The best way I found to improve my communication abilities is to volunteer to present at any opportunity, write blogs, and started making YouTube videos.
There are, of course, many other ways to improve your communication skills, but you have to actively seek them. Remember, the more you communicate, the more visible you become within the company. This increased visibility can open doors to new opportunities, including promotions.
Being visible is one of the reasons I was promoted; read more about that below.
Maths
Finally, we have reached the most important skill, maths.
If I were doing this survey, I would put maths as my top skill because things like Python and SQL are just tools you use to get what you want from the data.
To know what you want, you need to understand maths, which gives you the correct thinking and intuition about the problem you want to solve.
A question that I always get is:
How much maths do I really need?
The answer is pretty simple: you need an understanding of calculus, Statistics, and linear algebra at roughly a high school level, in my opinion.
This will be sufficient to understand things like gradient descent, Bayesian statistics, and eigenvalues used throughout the industry. Also, you are unlikely to actually sit down and do the maths; it's more important to better understand how machine learning algorithms work.
Of course, the more maths you know, the better, and it will allow you to go into more technical roles like staff or principal data scientists or even research industries.
For these types of positions, you would typically need a bachelor's degree in a STEM subject, ideally in physics, math, or statistics. A master's or PhD is even better.
But, for an entry-level position, the basics will serve you just fine, so don't worry too much about it!
I have an entire article explaining the maths you need for data science along with best resources that I will link below.
Summary & Further Thoughts
The skills you need to be a data scientist are relatively straightforward. However, the tricky bit comes in when you need to consistently try to learn them. I always think everything is quite simple; you need effort and patience to get the results you want!
Another Thing!
I have a free newsletter, Dishing the Data, where I share weekly tips and advice as a practising data scientist. Plus, when you subscribe, you will get my FREE data science resume and short PDF version of this AI roadmap!
