The world’s leading publication for data science, AI, and ML professionals.

3 Rookie Mistakes To Avoid When Building Your Data Science Portfolio

Here's what to do and what to avoid

Photo by Magnet.me on Unsplash
Photo by Magnet.me on Unsplash

Can you remember the last time you got a job just by submitting your resume? That’s right; the technical field has been restructured to a high revolutionized standard whereby employers no longer believe in a document (your resume) filled with experiences and grades from your educational background. They want to see what you can do or have done with the skills you possess.

The capability to show your would-be employers what you can do, instead of just telling, is very paramount to clinching every job opportunity.

Now here’s the catch, it’s not just about building a portfolio — you need to build one that will get you hired immediately. There are lots of beginner mistakes most data professionals make when building their portfolios. These mistakes usually disqualify them from lucrative data job opportunities.

In this article we will go through common mistakes often made when building a Data Science portfolio and how to avoid them. I have done research and personal communications with data science professionals (data analysts, product designers, data engineers, and others) about their experiences in order to simplify things. As well as the dos and don’ts of building a portfolio that sells. With that, let’s jump into it!

1. Creating a website

Except you are Andrew Ng, your website won’t get any traffic – especially if you are new to data science. No experience, no networks, and no connections. Of course, you can use a website as a portfolio, but why not use a platform where millions of like-minded data scientists are based?

It’s a win-win situation, you get ready-made traffic. Meaning more eyes on your projects and you get to display your skills and achievements to a wider range of employers.

Github is a popular platform for displaying diverse data science projects and experiences. For beginners looking to learn as well as grow their data science portfolio, Github is the preferred place for you to start. With a large number of technology enthusiasts, most especially data scientists, your work will easily be accessed by coworkers, project teams, and managers. I’m sure that beats growing a website with zero traffic. It’s better to channel that energy into making a mark among other professionals on Github. You’ll need it.

Let’s say you just concluded a project, and you would like to create a Github repository, it is very crucial that you upload a README.md file containing the title, synopsis, datasets, and project files.

A good README file should contain:

  • Your project’s title.
  • A detailed description of what it is about and the date of launch.
  • Steps used in the process.
  • How to run your code.
  • List credits and licenses.

Your README file should summarize in simple context what your project is all about, I like to make mine easy to read to both technical and non-technical individuals.

Most data scientists often leave links to their Github profile in the project/experience section of their resume. Employers can take a look at what you have built, the tools used, and how you did it. Bear in mind, to stand out among other experts in the field, you need to make your Github profile unique and professional in the best way possible.

2. Zero public evidence of your data science skills

Can you deploy codes? Visualize data into simple presentations? Create solutions to basic challenges in data science? There are lots of talented and gifted data professionals and from experience, the answers to these questions above will be a big yes. Then since you’re vast in your field, why aren’t you sharing your knowledge among data scientists with mutual interests?

Strong public evidence will speak numbers for you in reference to your data science skills. Lots of data professionals and enthusiasts often Google solutions when they get stuck with their analysis, and peradventure they got help through your works online, there Is a huge chance they would reach out to you and connect to discuss other opportunities. Networking is essential in technology.

Senior Data Science practitioner at DataCamp, David Robinson, narrates his experience with an employer who noticed the knowledge he shared online years ago. While discussing this in an interview on Mode Analytics blog, he shared how he got his first job in the industry without submitting his resume. He said,

The most effective strategy for me was doing public work. I blogged and did a lot of open source development late in my Ph.D., and these helped give public evidence of my data science skills. But the way I landed my first industry job was a particularly noteworthy example of the public work. During my PhD I was an active answerer on the programming site Stack Overflow, and an engineer at the company came across one of my answers (one explaining the intuition behind the beta distribution). He was so impressed with the answer that he got in touch with me [through Twitter], and a few interviews later I was hired.

While some of you might call this luck, this is a pure incident of,

hard work + patience = success.

Content creation has been one of the most neglected aspects of building a portfolio. Data Science communities, social media groups, YouTube, Quora, and microblogging platforms are important mediums you can leverage to share knowledge, build a following, and develop your very own online Data Science real estate.

With that, here are the best platforms you can use to build your portfolio:

Kaggle

Kaggle is the biggest data science forum in the internet space with various significant features required for data scientists to showcase their skills, share information on a wide range of data-related topics, build and publish data sets. There’s so much you can do with Kaggle to spice up your portfolio.

Participating in Kaggle competitions and creating useful models is a good way to show proficiency as a data scientist. Creating an online presence as a data scientist is very important if you want to build professionalism to the highest level, and Kaggle is a free community that will help you do just that at a high standard.

Twitter

Yeah, that same app you use to check if there’s an earthquake. The popular microblogging platform is a powerful tool for building a healthy data science portfolio. Twitter is a good place to recognize and interact with other professionals in your field. Companies and individuals share opportunities (job offers, freelancing promotion, and invites to conferences) on their Twitter page as well.

Tableau Public

Tableau is a data visualization software designed to convert bulky raw data into useful and readable information by presenting them in form of graphs and charts. Most data jobs require you to have basic knowledge of Tableau before you can receive an offer. Displaying a few dashboards on Tableau Public will add credibility to your skills.

I recently published an article on the best features of Tableau that will improve your data science and visualization skills. Feel free to check it out to get a detailed understanding of the Tableau software.

3. Including the wrong projects

There are lots of projects to work on in a wide data science niche, it is worth mentioning that knowing what kind of projects to add to your portfolio is a very important skill every data scientist needs to know and apply with precision. Machine learning, Visualization, IoT, Artificial Intelligence, the list goes on and most times it’s challenging to know what kind of projects hiring managers want to see.

William Chen, a Data Science Researcher and interviewer at Quora described the categories of unique projects on every Data Professionals portfolio:

I love projects where people show that they are interested in data in a way that goes beyond homework assignments. Any sort of class final project where you explore an interesting dataset and find interesting results… Put effort into the writeup… I really like seeing really good writeups where people find interesting and novel things…have some visualizations and share their work.

It’s really easy for employers to get turned off with the wrong project — displaying the perfect projects will get you closer to your dream position.

Very important: It might be tempting to throw all your Data Science work straight into your portfolio. Identify the projects that display your skillsets to the fullest. Try including that work you did for your biggest client, a complex but simplified thesis, that project where you created a solution to a popular problem.

Conclusion

A strong resume will get you into the office as for the interview, a strong portfolio will get you into the office as a fully qualified data scientist. The important thing is to get involved in more projects with professionals with higher experience. I always tell my colleagues: no matter what you think you know, you can never learn everything about data science. Keep learning and building.

As David Robinson phrases it,

Generally, when I’m evaluating a candidate, I’m excited to see what they’ve shared publicly, even if it’s not polished or finished. And sharing anything is almost always better than sharing nothing.

The more projects you engage in, the more skills you can showcase in your portfolio. As you grow, make sure you continue to update your portfolio. Never stop sharing knowledge, you’ll increase your visibility and chances of being given an offer even without applying.


Resources

http://varianceexplained.org/r/start-blog/


Related Articles