KDnuggets Top Blog Winner

A Data Science Portfolio That Will Land You The Job in 2022

Check out this article on crafting a data science portfolio that will get you that job. And learn 4 resume mistakes to avoid at any cost.

A Data Science Portfolio That Will Land You The Job in 2022
Image by Editor


If you are reading this article, it is likely that one or more of the following statements apply to you:

  1. You do not have a Master’s degree in data science and have no prior experience in the field.
  2. Every data science job posting you come across requires at least 2-3 years of experience in the field. But how do you gain experience if nobody will hire you without any?
  3. You have taken countless data science online courses, and they all teach you similar things. You end up getting stuck in a trap where you aren’t really learning anything new, you just feel like you are.
  4. You have even created data science projects and included them in your portfolio hoping that they will get you a job in the field, but they didn’t.

If any of the above statements resonate with you, here are 4 things you are probably doing wrong:


1. You Are Taking Too Many Online Courses


Every time you take a new online course that’s made for beginners and add it to your resume, you are telling employers that you are new to the field. This makes it seem like you lack experience, and can do more harm than good to your portfolio.

I’ve seen many data science candidates list over ten similar online courses on their resume with no projects or real-world applications that demonstrate the skills they learned.


Here’s what you should do instead


Take one or two data science online courses and create projects based on the skills gained. If you find that you lack expertise in specific areas while building these projects, then take a course that bridges the gap in your knowledge.

This way, you not only learn faster but can display a wide variety of skill sets on your resume instead of listing ten courses that teach the same thing.


2. You Don't Stand Out From the Crowd


Another mistake most applicants make is that they list the same types of projects on their resumes.

Projects like Iris Flower Classification and Titanic Survival Prediction are extremely popular. 

Due to their simplicity, these are the first projects most people create when learning data science. If you add them to your resume, you are telling employers that you lack experience and creativity.

Hundreds of other applicants have created the same project and mentioned it in their portfolios. Why should you get the job offer when your resume is no different from 80% of other data science aspirants out there?


Here’s what you should do instead


Create data science projects that showcase a variety of skills, such as data analysis, machine learning, and data preprocessing. If you need inspiration to create unique projects that stand out, take a look at 5 of my best data science portfolio projects.

Real-world data is often dirty and needs to be collected from external sources, unlike Kaggle datasets that are already structured and preprocessed.

It is a good idea to create an end-to-end data science project that involves collecting data through an API or web scraping techniques. 

This shows employers that your programming skills are strong enough for you to work on real-world data science use cases.


3. You Are Not Doing Enough Research


Many job seekers tend to cold submit their resumes to every open data science job posting, thinking that it will increase their chances of landing employment.

However, doing this actually reduces your odds of getting a job. It is likely that you will either get a generic rejection email or no response at all from the companies you applied to.

This is because you are applying without considering what the company is actually looking for. Every open position is unique, and different companies hire data scientists for different reasons.

For instance, an eCommerce company might hire a data scientist in their marketing department to build a recommender system that encourages customers to make more purchases on the site. 

On the other hand, a tech company may hire a data scientist to help their product team introduce new features and measure product success.

While the job title is “Data Scientist” for both of these roles, their job scopes vary. If you apply to each job posting with the same resume without taking the company’s use case into consideration, it shows.


Here’s what you should do instead


Select a handful of companies you want to apply to. Read about them and do some research on the kind of industry they operate in.

Then, try creating projects that are relevant to this industry. This tells hiring managers that you will add value to them since you have worked on projects that are similar to their use cases.

You can even go a step further and reach out to data scientists who already work in the company you want to join. Connect with them on LinkedIn or via email, and try to get an idea about the kinds of projects they are involved in.

You can then create something relevant so that your resume stands out among other applicants.


4. You Are Not Playing to Your Strengths


A few months ago, an aspiring data scientist reached out to me. She wanted to land a job in data science but was unable to do so.

I reviewed her resume and realized the problem immediately.

This applicant came from a marketing background and had only been to one data science Bootcamp.

The skills she had highlighted on her resume included programming, machine learning, and statistics.

Any employer reading that resume would have been able to tell that her knowledge in the above subjects was limited since there is only so much you can learn from a 3-month Bootcamp.


Here’s what she should have done instead


This applicant’s strength lay in her marketing domain knowledge.

With a background in marketing, she should have created projects relevant to marketing analytics.

Most marketing experts cannot deal with large amounts of data. They lack technical and analytical skills.

Since she already possessed domain expertise in marketing, all she had to do was pick up some Python, Excel, and SQL skills. Then, she should have created a few marketing data analytics projects using these tools and included them in her portfolio.

This would have easily landed her a job in analytics. 

Then, after one or two years of experience in the field, she could have made the transition from data analytics to data science.

Data science is a field that requires knowledge of applied statistics, along with strong analytical and programming capabilities. For most people, it is difficult to get a data science job right away without formal qualifications. 

Due to this, it makes sense to first break into a field with a lower barrier of entry like data engineering or analytics, then gradually make the transition into data science.


How My Data Science Portfolio Landed Me a Job in the Field


I’ve landed multiple roles in the data science field without a Master’s degree. 

After my first data science internship, I became a senior consultant at the same company, and am now working as a forecasting analyst at an MNC.

I got these jobs not because I was the most qualified candidate applying to these positions, but because I demonstrated a genuine interest in the field. Employers look for applicants who are passionate about the work they do, since skills can be developed over time.

I created data science projects on topics I was interested in and wrote blog posts about them. I published my work on my data science blog and portfolio website.

Then, I included links to my website and blog on my resume. I highlighted some of my best projects in the “Projects” section of my resume and wrote a few lines about them.

After getting my first actual data science role, I gained some experience in the field, which helped me create bigger and more interesting projects. As I continued to publish my work online, employers started reaching out to me for freelance projects

I was hired to collect large amounts of data for companies, build machine learning models, perform market research, and engage in freelance writing projects. All these gigs were on top of my day job as a data scientist.

Establishing a strong online presence is the key to landing job opportunities. There are many skilled people out there who never get discovered because employers do not know where to find them.

Natassha Selvaraj is a self-taught data scientist with a passion for writing. You can connect with her on LinkedIn.