Taylor Rodgers
  • Home
  • Books
  • Free Tools
    • Requirement Doc Templates
    • Quality Checking Guide
  • Blogs
    • Data Science
    • R Programming
  • Consulting
    • Statistical Consultant
  • Contact
  • Home
  • Books
  • Free Tools
    • Requirement Doc Templates
    • Quality Checking Guide
  • Blogs
    • Data Science
    • R Programming
  • Consulting
    • Statistical Consultant
  • Contact

How I Finally Became a Data Scientist – And How You Can Too

2/12/2021

0 Comments

 
Picture

On December 21st, 2020, I started my dream job. I had finally made it into the most prestigious role in data and analytics – I was a bonafide data scientist.

It’s funny because one year and two months earlier, I thought about leaving analytics for good. I found myself stuck in the reporting trap – where it seemed like the only thing I could ever produce was another Tableau dashboard – despite my best efforts to move towards analysis work.

I didn’t feel like I had control over my career. It seemed like other people made those decisions for me. Every time I did something outside dashboard work, I’d hear the most horrific words you could ever hear at work: “that’s not really your job to worry about.”

And so I drew a line in the sand. I said no more. I got in this field to use statistics and programming to find cool insights and make real world predictions. That’s what I’m going to do.

I left my job – one that paid well for easy work – without another lined up – a move that concerned family and friends alike.

I devoted myself to finding the data scientist role I always wanted and managed to finally do it one year later.

Many Smart People Never Become Data Scientists – Why Did My Story Turn Out Differently?

It’s hard not to feel like one of the lucky ones. Many other people aspire to become data scientists. It’s not uncommon to see openings for data scientists to have hundreds of applicants.

That’s understandable. On average, data scientists make about $120k a year. The work-life balance is generally pretty good and the work is intellectually rewarding.

After landing my dream job, it didn’t take long for other aspiring data scientists to contact me on LinkedIn to ask me – ”how can I land this sweet gig?”

Well… it depends on who you ask. I knew many other data scientists and I had asked them all the same thing. There were some overlaps in their answers, but there was also a lot of variation.

Some created project after project with Kaggle datasets. Others offered to work for free for existing data science teams. Others simply started doing the work of a data scientist at their current job and got promoted. And others just applied and got hired right out of the gate! Without a graduate degree no less!

And that’s the harsh truth about data science. Every data scientist has a different story for how they landed their dream job. As much as I wanted to find the silver bullet – the one true path guaranteed to get me a data scientist job – there really wasn’t one.

I had to find my own path and develop my own strategies. And it’s my hope that you learn from my journey with your own job search.

My Strategies for Becoming a Data Scientist

In the year between my last job and my new job, I had three broad strategies:
  1. Demonstrating aptitude to potential employers through graduate school and side projects
  2. Using my best medium (writing) to develop a reputation among the data science community
  3. Good ol' fashioned networking

Just because the steps are simple and straightforward doesn’t mean they’re easy. I had to apply myself academically – for the better part of a year – and take significant financial risks – all on the chance that my plan would work.

Let’s go into more detail about how I used these strategies.

Strategy #1: Demonstrating Aptitude

The best jobs usually require several years experience. That’s particularly true with data scientist jobs. The catch-22 is that you don’t have the experience for those jobs until you land those jobs.

Seems like a rigged system, doesn’t it?

I had several years experience in data and analytics. Less so in data science. To industry outsiders, there doesn’t seem like much of a difference, but there really is. One focuses on database and report automation. The other focuses on predictive analytics, machine learning, and statistical inference.

A couple years ago, I had a phone call with an HR rep for a data scientist position. She listed off the job requirements and said: “After looking at your resume, I frankly just don’t see it.”

I told her I may not have the work experience, but I certainly had the aptitude. I listed off a few statistical analyses and machine learning work I had accomplished. She didn’t buy it.

That call frustrated me, but I learned an important lesson. The gatekeepers to data science roles prioritized experience. I didn’t have the experience they wanted. That meant I had to work far harder to demonstrate the aptitude to overcome that barrier.

And how can people without professional experience demonstrate aptitude? There are two ideas that come to mind:
  1. Education
  2. Side projects

How a Graduate Degree Demonstrates Aptitude for Data Science

The biggest aptitude indicator for data science is a master's degree. Data scientist positions used to require PhDs. Now they require a master's degree, or at the very least prefer one.

Now I’ve known several data scientists without masters degrees. They’re the exception to the rule and most of them attend graduate school after landing their job anyways. Not only to learn new techniques, but because they know that even with professional experience, many employers won’t even bother calling them without a graduate degree.

This is different from the rest of the programming profession. It’s not uncommon for those without a college education at all to work as a programmer. It’s a badge of honor in some circles!

It’s harder though for data science. Mostly for the simple fact that it’s hard to teach yourself statistics via blog posts and books. You have to know calculus and a little bit of matrix algebra for starters, which you usually learn in college.  You also need guidance on what areas of statistics to study.

Many people who claim to know data science only know simple concepts, such as “decision trees.” These methods are handy to know, but probably the most elementary of techniques. There are more complicated ones to master.

I thought I could learn these techniques on my own. I took two statistics courses in my undergraduate. I felt I had a strong grasp of regression analysis and, to a lesser degree, logistic regression.

However, I kept running into problems with many data sets. The data simply didn’t match those techniques and I didn’t know the right terms to find other methods to apply.

Graduate school helped with this tremendously.

Multivariate statistics, survival analysis, analysis of variance, and categorical data analysis are all techniques that they don’t teach in significant detail in your undergrad. And I can’t imagine I ever would’ve discovered them on my own through web searches. I’m sure others have, but a graduate degree was worth it for me.

The other added benefit with a graduate program is I learned the language of data scientists. All the buzzwords and jargon employers use to intimidate and reject applicants became familiar. Concepts with complicated sounding names turned out to be far more simpler than I expected.

That helped me write a far better resume and feel more confident that I could speak to key concepts in job interviews.

How Personal Projects Demonstrate Aptitude

One data scientist I spoke to who landed a data scientist job prior to finishing graduate school told me she worked on side projects. Whenever she went in for an interview, she had a whole portfolio to present!

That’s not a bad approach, I’d say.

I attempted to find my own projects. I didn’t want to use kaggle datasets though. I wanted to solve a real-world problem I came across. Sadly, I had trouble with that.

There were two big ones I thought had potential. One related to potholes and wheel and axle damage near my house. The other related to police stops that an attorney (a family member of mine) believed was unconstitutional.

Those projects fell through for different reasons.

I later settled on a different project – the R Programming in Plain English book and blog. I had worked with R programming for several years and after taking a class in my graduate program this past summer, I realized I could explain it better.

This worked well because R programming – along with Python – is the primary programming language for data science.

Writing and publishing articles on R programming was a great thing to include on my resume. Whenever I applied for jobs, I literally included links!

It’s pretty hard to say someone isn’t qualified when they literally wrote a book on the subject. :)

Writing is an example of using a medium to show the world your passion and expertise, which segues nicely into my next point…

Strategy #2: Finding My Medium and Using It Well

The best career advice I got was about finding the right medium. That medium could be writing, public speaking, podcasts, youtube videos, etc.

Learning your best medium allows you to demonstrate your passion for a career to many people, such as future colleagues and employers. It’s also handy for demonstrating aptitude as well.

Writing is my primary medium. Since 2019, I’ve written many articles on data science and analytics. These have collectively over 20k views and something I regularly use to meet well-respected individuals in my career field.

Public speaking is a medium I’ve used to a lesser extent, but it has brought me job leads.

About a year and a half ago, I spoke at the Kansas City Tableau User Group. I worked really, really hard to make the talk as fun and engaging and different from other talks I’ve seen before. Afterwards, I had multiple people reach out to me to see if I was interested in working with them.

If you put a lot of effort into making a fun, engaging, and informative talk, you could experience the same thing. It'll introduce you to many new people and demonstrate your aptitude and talent to those inclined to hire you.

Whether it's public speaking or writing or podcasting, find the best medium for you and use it extensively. You'll seldom make money, but it'll make you more attractive as a candidate and helps form new relationships.

Strategy #3: Building Relationships

Most people are bad at networking. It turns out, I’m fairly good at it. The reason is that I don’t think of it as networking, but as “relationship building.” You’re connecting with people that have the same passion for your career field as you. (You can read about my networking approach here.)

In the past year, I spoke with well-known authors, executives, and other top-performers in the data science biz. I never did this with the intention of getting a job though. I merely wanted to connect with them. (One caveat: I met some to interview them for my articles.)

Several of these conversations led to strong job leads. It wasn’t the direct goal of networking – it was a happy byproduct. That’s how I suggest you approach networking as well.

In marketing terms, it’s like building brand awareness among your professional peers.

Elevator Pitches Help When Building Relationships

Almost every individual I met had a unique approach to their career. They could usually define it in a few sentences. This is commonly called an “elevator pitch.”

I learned to share my own elevator pitch with these folks. I wanted to make it known what my career specialization was all about. That way, when the right opportunity that aligned with that focus came along, I’d be top of mind for it.

I told them, “I want to be the guy that takes the complicated things about data science and makes it easy-to-understand. I want to make it more accessible to a general audience.”

This elevator pitch worked. Consulting agencies in the data science space, which prioritized client-centric communication, were always the most interested in hiring me.

How These Three Strategies Led to My First Data Science Job

I had many job leads and many led to dead ends. But I only needed one and that happened in November, 2020. About one year and one month after I left my previous job.

Here’s how I used my job hunt strategies to land it.

I finished 27 out of 30 credit hours of my masters program, which demonstrated aptitude.

I wrote and produced digital content about data science, finished writing my first book’s manuscript, and started writing my second book. In other words, I used my medium to build credibility.

I also met and spoke with various people in my field. That was relationship building. I met one in particular at an R User Group. He gave a talk there and I reached out to him afterwards. We met for coffee a few times and, after the pandemic hit, spoke over the phone.

I originally reached out to him to find a data science case study to write about. I wanted to get it published in the Harvard Business Review to explain what a successful data science solution looked like since many business leaders frequently misunderstand its use cases.

Most private companies want to keep data and insights private to remain competitive. His company did pro bono work for a non-profit though. That sounded like a good case study candidate to me. A non-profit charity, unlike a private business, would be more open to publicizing a case study like that.

That project never panned out for other reasons, but my contact and I built a good relationship during our talks. He later recommended me for an open data scientist role on his team – the position I later accepted.

During the hiring process, they told me that my graduate education made me more attractive as a candidate and that they loved my client-focused communication style. In other words, my elevator pitch resonated with them.

Final Thoughts – It’s Not Always Smooth Sailing

Having a career is a lot like sailing a boat. You can’t sail straight for your destination. You’re at the mercy of the winds, the weather, and the tide. Despite how hard you work, sometimes factors outside your control make it impossible to reach your destination when you’re ready for it.

That’s how my journey felt. On the outside, it may have looked like I knew what I was doing the entire time. I earned an education, wrote a book, produced content, and made various connections and conversations with industry experts.

It did not always feel like smooth sailing though. Even after I accepted a job offer, I didn’t think it was real. I was even afraid to update LinkedIn. I thought it’d somehow be taken away from me because of the many setbacks I experienced that year.

The biggest setback was the pandemic. It affected me like it did others.

Multiple times, I thought I had landed the breakthrough role. I’d get a warm reception only to later hear “we’ve put things on hold right now.”

Other employers specifically set a start date and later ghosted me!

But it finally happened. I reached port and landed my dream job. Life had never seemed sweeter. :)
0 Comments



Leave a Reply.

    ABOUT

    A blog about the non-technical side of data science.

      SUBSCRIBE

    Confirm

    ARCHIVES

    April 2022
    March 2022
    October 2021
    September 2021
    August 2021
    March 2021
    February 2021
    January 2021
    November 2020
    August 2020
    June 2020
    May 2020
    April 2020
    February 2020
    January 2020
    December 2019
    November 2019
    October 2019
    September 2019
    August 2019
    July 2019
    June 2019
    May 2019
    March 2019

    RSS Feed