How to be a Data Scientist without a STEM degree


1. Learn the fundamentals of all pillars of data science

“Data Science” is a vague term—it can mean different things to different companies, and there are a plethora of skills that are relevant to data scientists.

That being said, there are a few core skills that I recommend that you learn. The following skills are pivotal for any data scientist: SQL, Python, Statistics, Machine Learning. I also recommend that you learn these skills in that order. It may sound like a lot, but it’s no different than when you had to complete 4–6 courses per semester in college!

Let’s dive into each skill.

 

A) SQL

 

SQL is the language of data and is arguably the most important skill for any data scientist. SQL is used to manipulate data, analyze data, build dashboards, build pipelines, write queries to feed into models, and the list goes on.

 

B) Python and Pandas

 

Python (or any scripting language) serves as a foundation for doing several other things like building ML models, web scraping data, building automated scripts, and so forth.

Pandas is a Python library used for data manipulation and analysis. I personally use Pandas over SQL when exploring data in a Jupyter notebook.

Below are the most useful resources I’ve used to learn Python and Pandas:

 

C) Statistics

 

Data science/machine learning is essentially a modern version of statistics. By learning statistics first, you’ll have a much easier time when it comes to learning machine learning concepts and algorithms! Even though it may seem like you’re not getting anything tangible out of the first few weeks, it will be worth it in the later weeks.

Below are the most useful resources I’ve used to learn Statistics:

 

D) Machine Learning

 

Not only is machine learning interesting and exciting, but it is also a skill that all data scientists have. It’s true that modeling makes up a small portion of a data scientist’s time, but it doesn’t take away from its importance.

Below are the most useful resources I’ve used to learn Machine Learning:

2. Complete 1–3 personal data science projects

Once you have a foundation built, the best way to accelerate your learning is by completing some data science projects. The simplest way to do it is to go on Kaggle, pick a dataset, and create a prediction model or some data visualizations. Remember that your first few projects aren’t going to be great! But what matters is how you progress over time.

Here are some data science projects that I completed in the past that you can use to get some inspiration!

While you continue to learn and practice your data science skills, there are other things that you can do to make yourself a more valuable data science candidate, and this leads to my next tip.

3. Explore unconventional opportunities for experience

The hardest part of being a data scientist is getting your first opportunity with no prior experience. However, below are several ways that you can get experience even if you don’t have experience:

 

Non-Profit Opportunities

 

Recently, I came across a resourceful article written by Susan Currie Sivek, which provides several organizations where you can find opportunities to work on real-life data science projects.

If you’re trying to find more experiences to add to your resume, I highly recommend that you check this out.

 

Compete in competitions

 

In my opinion, there’s no better way of showing that you’re ready for a data science job than to showcase your code through competitions. Kaggle hosts a variety of competitions that involve building a model to optimize a certain metric.

Two competitions that you can try right now are:

  1. Titanic: Machine Learning from Disaster
  2. House Prices: Advanced Regression Techniques

 

Start a blog on Medium

 

Yes, I’m biased, but hear me out. You’d be surprised how many data-related professionals are on Medium. They like to see informative, insightful, and interesting material. Take advantage of Medium to blog about your learnings, to explain a complex topic in simple jargon, or to walk through your data science projects!

Specifically, I recommend that you write for the publication Towards Data Science, as they currently have a follower base of almost 500,000 followers.

If you’d like some inspiration, check out my project walkthrough on Wine Quality Prediction.

4. Look for jobs similar to Data Scientist positions

I knew I would be fighting an uphill battle, especially with no previous experience as a Data Scientist. However, finding jobs similar to data scientist positions will significantly increase your chances of becoming a data scientist. The reason for this is that related jobs will give you the opportunity to work with actual data in a business setting.

You don’t need to be a Data Scientist to do ‘Data Science’ work

Here are some data science-related jobs that you can look for:

  • Business Intelligence Analyst
  • Data Analyst
  • Product Analyst
  • Growth Marketing Analyst / Marketing Analytics
  • Quantitative Analyst

In addition to the two points above, there’s one more tip that significantly improved my reputability as a data scientist.

5. Consider getting a Master’s degree in a quantitative field

Most Data Science job listings require a Master’s degree because it generally requires a high level of technical skill. If you find that you are not finding success with the two pieces of advice above, I recommend looking into a Master’s program in a quantitative field (computer science, statistics, math, analytics, etc.).

Personally, I chose to enroll in Georgia Tech’s Master of Science in Analytics program for a number of reasons:

  1. It doesn’t require a bachelor’s degree in a quantitative field.
  2. It has an online program in case you want to work and study at the same time.
  3. It costs only $10K USD for the whole program.

That being said, there are several options out there, and I highly advise that you take the time to explore all of your options before you make a decision!

Related:



Source link

Leave a Reply

Your email address will not be published. Required fields are marked *