How to Jumpstart Your Career In Data Science

How to Jumpstart Your Career In Data Science

Making the jump from aspiring Data Scientist to full-fledged, knows-what-they’re talking about Data Scientist is pretty hard. You have to know the statistics, algebra and have some business sense. Thankfully I’ve interviewed at a lot of companies hiring for Data Science roles and got a lot of valuable feedback. I’ve distilled that feedback into this short list with resources so you can up your game and land the job. 🖥

1. Highlight your passion projects on your resume. 🚀 This is important for folks either switching careers or without a full time “Data Scientist” time on their resume already. The focus is to show off how you work, what concepts you understand, and how they apply to a company. When you get to the interview stage expect to go in depth about why you chose one algorithm over another, hyperparameters you tuned (or didn’t) and what your results were. I urge you to practice this with a friend because interviewers may not care as much about if you chose random forests over decision trees, but they care about why you made that decision and how well you can explain it. Kaggle is an amazing resource of example projects to get you started. If you like defining projects yourself, Data.World has plenty of datasets to dig into.

2. Get comfortable using git 📟 Whether you’ll be at a startup or a big corporation, chances are that you’ll be working with a team of developers and data engineers. Understand how good version control can benefit your team and to get familiar with it, 🔊 USE GIT FOR YOUR PROJECTS. You’ll be happy when you can figure out what went wrong when you revisit your code and when you start that big job you’ll be comfortable with Git, instead of feeling overwhelmed with another technology to learn. freeCodeCamp has created my favorite short intro to git.

3. Teach other people 👩🏾‍🏫One of the best ways to really “chunk” a concept into your brain is to teach it. By having to explain concepts in different ways, you’ll be forced to understand it deeply. Whether this is talking to your family, teaching at a code bootcamp, or talking in a Meetup, you’ll soon notice how much better you understand a topic.

4. Practice your SQL skills 🗂 Almost all organizations use SQL or a variant like MySQL or NoSQL to store their data. Show hiring managers you’ll be independent to get the data yourself by letting those SQL skills shine. While it’s fairly easy to learn, I’ve found retaining my SQL skills has been difficult if you don’t use it every day. Create a free database using something like SQLite to start managing some of the data you care about in a secure place. I’ve done this for my marketing analytics and even college grades. The point is to practice enough so you can ace the data retrieval part of your interview. Mode Analytics has a fairly in-depth SQL tutorial recommended by recruiters at the big 4.

5. Know your stats 📊I’ve been in many interview situations where they want you to calculate some statistic on the spot. They usually aren’t testing if you have the Wolfram Alpha definition memorized, but moreso to see if you can be their occasional statistics go-to. They also don’t expect you to know everything (unless you're applying for a 100% Statistician role) but you do need to know about statistical significance, hypothesis testing, p-values, distributions, and possibly A/B testing. I think the best resource for getting this is Udacity’s FREE Intro to Statistics Course.

6. Learn how to communicate well 🗣 If you want to be the badass expert on your team then people will go to you for questions. Your manager probably won't expect you’ll do these tests in a vacuum and never talk about it. In a behavioral interview, a good method for organizing your response is to use the STAR method. Talking over the situation, task, action, and result help to clearly explain the instance you’re describing.

7. Get comfortable reading documentation 📖 It may seem like a really boring task and trust me, the technical writers aren’t bursting at the seams with humor, but these documents are so valuable to understanding why something isn’t working. Don’t look at viewing documentation as not knowing what you’re doing. Get used to being unsure about what’s going on sometimes; the only “official” way to check what’s going on is with documentation. I keep a folder in my bookmarks for the documentation of libraries I use regularly. You have no excuse when the answer is at your fingertips.

8. Define what you want in your next role 💎 There are thousands of Data Science openings, but not each role is alike. For new Data Scientists, large companies tend to provide security and a mentorship focused culture. On the other hand, startups can provide more project ownership and responsibility. It’s good to know about the unique problems your target company may face, as well as industry-related issues. Discover the types of issues you might face in each vertical and think about how you might deal with them day to day. The best way to learn is to talk to the practitioners. There are some great podcasts that highlight practical applications of artificial intelligence in different industries. You can use a tool like Fairygodboss, Glassdoor, or KeyValues to find out a company’s internal culture.

Remember, you’re working to align yourself to be the best choice in a highly competitive field. Data Science is extremely interdisciplinary so you’ll have to build on your soft skills as well as your technical ones. I hope you find these resources useful to helping you stand out in your next interview.

 What does Focal Loss Mean for Training Neural Networks

What does Focal Loss Mean for Training Neural Networks

5 TV Shows to Watch While You Code

5 TV Shows to Watch While You Code

0