Candid Advice for Aspiring Data Scientists

Candid Advice for Aspiring Data ScientistsHow to become a Data Scientist and land your first jobClaire LongoBlockedUnblockFollowFollowingMar 5Photo by MD Duran on UnsplashMany people are interested in becoming Data Scientists, but it can be difficult to identify what you need to learn to land your first Data Science job.

These tips are for anyone who is starting their Data Science career with the goal of obtaining a job in industry and applying Machine Learning and Deep Learning to solve business problems.

Ok, so what do I need to learn to be a Data Scientist?If you start with the list below, you will be prepared to tackle just about any Data Science task you’ll be faced with on your first job.

Python and/or R (The core machine learning libraries in Python are pandas, numpy, scikit-learn, seaborn, and matplotlib)A deep learning framework like tensorflow or pytorchStatistics (hypothesis testing, statistical modeling, probability, bayesian statistics)Machine Learning (unsupervised modeling and supervised modeling)Deep Learning (neural networks)Github and virtual environments for version controlHow to deploy a machine learning model (You can use a machine learning platform like SageMaker or Domino, or you can do it yourself with Flask in Python)For Statistics, Machine Learning, and Deep Learning, you will need to understand the mathematics of the models, be able to explain the model intuitively and implement the models in Python or R.

Through continued job experience, you’ll refine your Data Science skills and develop your own best practices for things like feature engineering, evaluating models, model monitoring, automated retraining, avoiding mistakes like data leakage, and presenting your work to a broad audience.

Do I need a degree?The answer is no, but it does help if you have at least a Bachelors degree in a quantitative field.

The best way to learn Data Science depends on your learning style.

Degrees, immersion programs, and self-teaching with online courses are all viable options.

The most important thing to do is identify how you learn, and then choose a learning path that works for you.

If you need structure and deadlines to stay motivated and on-track, opt for a degree or immersion program.

If you already have a strong mathematical background, I recommend the online course route because it is cheaper, and prepares you very well for the job.

Beating imposter syndromeImposter syndrome is everywhere and Data Science is no exceptions.

I have two methods to overcome this.

Choose a focus area to master.

The field of Data Science is still being developed and defined, and as a result, there will always be something new and crucial to learn.

It can feel daunting and frustrating to tackle learning the entire field of Data Science at once.

I recommend developing a broad knowledge in all the methods and tools listed above, but also choose an area within Data Science to focus on and master.

For example, many Data Scientists make NLP or Deep Learning their focus.

Pick something you really enjoy or something you could use to solve problems in a field that you’d like to work in.

Selecting a focus area allows you to truly master a small, digestible piece of the field while still maintaining your working knowledge of the entire field.

Becoming an expert in one area will build your confidence in the field, and can help you establish yourself as an experienced Data Scientist.

Talk about the stuff you don’t know as well as the stuff you do.

As Data Scientists, it is important to recognize and discuss the things we don’t know, as well as the things we do.

If we have open conversations about topics that interest us that we’d like to learn more about and discuss mistakes we’ve learned from, we will be able to learn from each other and ultimately become better Data Scientists.

But if we pretend we’ve mastered everything and never make a mistake, we’ll make others feel inferior and discouraged, and will not be able to learn from those around us.

If you find yourself working in an environment where you feel like you can’t make a mistake, or can’t admit that you don’t know something, you’re in a toxic environment and you should probably start looking for your next job opportunity.

The importance of a portfolioYour portfolio is crucial if you are getting started in your Data Science career.

You will need some project work to discuss in the interviews.

Both personal projects and Kaggle competitions are great ways to build your portfolio.

Kaggle has some great free datasets you can download and use for a personal project.

Personal project more accurately represents a real-life project because it requires some creativity to define the project, choose the modeling approach, and choose the right evaluation metrics.

Having a Github page with personal projects will be a huge advantage during your job search.

Many employers will look at your Github prior to an in-person interview to see a sample of your code.

It will not help you to just have some notebooks from online coursework on your Github.

I recommend creating repos for your personal projects and writing up a simple description of the project in the Readme.

How to land that job!Now you’re ready to start interviewing! I don’t know the secret recipe to get your dream job, but I do have a few interviewing tips that I believe will increase your chances, and put you towards the top of the stack of applications.

Study the business.

Come to your interview with a list of ideas of how the company could use Data Science.

I’ve done this multiple times, as well as observed others do it, and I believe it was the key strategy that ultimately leads to a job offer.

Be energetic and positive in the interview.

It’s easy to be serious and focused when you’re nervous, but it is important to still appear excited about your work.

If you can try to connect with your interviews and be conversational, you will make a good impression, and you may become less nervous as the interview will become more of a conversation and not an interrogation.

Treat interview questions as talking points.

This is especially important in phone screens.

Early in my career, I made the mistake of answering the questions with a simple sentence, or even just “yes” or “no”.

This was totally the wrong approach.

Take the opportunity to explain your answers in detail, or even use the question to jump into something you want to talk about.

Have opinions about the field of Data Science.

Do you have ideas on how Data Science should or shouldn’t be used? What makes a Data Science team successful? Have you seen something derail a project in the past? How should a Data Scientist even be defined? These are all insights about your field that you could bring to the table.

Because the field is so new, we all have the responsibility as Data Scientist to help define and clarify what a Data Scientist does.

I thought I should keep my opinions to myself until I realized that because Data Science is so sought after and yet still so difficult to understand and define, most companies actually wanted to hear what I thought.

Know everything on your resume.

If it’s on your resume, its fair game for an interviewer to ask you to describe it.

Don’t put anything on your resume that you can’t explain or remember.

Before interviews, I practice describing all the models I’ve used on past projects because I anticipate direct questions related to these models.

With these tips, you should be able to learn everything you need to know to start your career as a Data Scientist and stand out in the interview process.

I hope these tips will make the process of learning Data Science easier.

Disclaimer — This article contains my own candid advice and all information in this article is my own opinion based on my personal experience as a Data Scientist and is not scientific fact.

If you disagree with any of the advice in this article, feel free to leave a comment explaining why.

If any of this info helped you out, I’d love to hear that as well.

 :-).. More details

Leave a Reply