4 Ways to Get the Most Out of Your Data Science Degree
It was March 2021, and I'd just received an email which was about to change the course of my life.
"I am writing to inform you that we would like to make you an offer for 2021 entry to the University of Oxford."
I think I squealed a little. Or did a silly dance. Maybe both.
Getting into Oxford was a huge deal for me. I'd known for a while that I wanted to change careers into Data Science, and unlike some of my coursemates, I wasn't coming straight from my bachelor's degree. I'd already been working for a few years since graduating, and I was entering this field with two very specific aims:
- Enjoy it
- Set myself up for working as a Data Scientist in industry
As I quickly found out, however, juggling these two aims in your degree is quite a difficult balancing act. In this article I'll share 4 things that worked for me in the hope that they help you, too.
Write your thesis in collaboration with a real-life organisation
Academia is full of people doing super interesting research.
The problem is that a lot of research is very theoretical, and if you're aiming to go into industry after completing your degree, a lot of academic research projects won't be particularly relevant to the employers you want to work for.
During my bachelor's degree, for example, I poured my heart and soul into a 10,000 word dissertation on the foreign aid policies of Arab countries. I loved the project, but aside from the soft skills I honed through it (communication, research, etc.) it had very little relevance to any of the jobs I could get afterwards. The project gave me a lot of life experience, but in terms of my employability, I think it added very little.
As a result, when I started my master's degree, I knew I wanted to take a different approach. Each year, my academic department at Oxford would send out a 30+ page booklet listing potential thesis projects we could work on, yet I knew right away that none of these ideas would work for me. My main aim in the degree was not to contribute to some esoteric academic project. It was to prepare for working as a Data Scientist in industry, and I knew that doing an academically interesting yet commercially irrelevant project wouldn't help me towards that aim.
Don't get me wrong – I've got nothing against niche academic projects. Life's about more than economic output and careers, and it enriches society to have people doing research even if it isn't directly "industry-relevant." But because I knew I wasn't going to be sticking around in academia for long, I decided that my efforts would be best spent on trying to find a project which was directly relevant to an industry use case. If I could find one like that, I figured, the thesis could double up as commercial(ish) experience that I could talk about on my resumé and in job interviews. This would be especially valuable for me, given that I didn't have any prior commercial experience in Machine Learning/Data Science.
And so, when the time came to start planning my thesis, I started looking for an industry-relevant project. How did I find a project like this? I cold-messaged 5 people on LinkedIn who worked in Data Science/ML teams at cool companies, mentioned a couple of ideas about how I could contribute to some problems they might be working on, and asked if they'd be up for collaborating with me on a Data Science project. Why only 5 messages? Because that's the limit for a non-paying, non-Premium LinkedIn account!
From my past experience working in Outbound Sales, I knew not to expect a high response rate. The people I messaged were senior members of their organisations, and I knew that they might not have time to help me with this kind of request. But that didn't deter me: all I needed was one person willing to take a chance on me.
Eventually, I got a bite. A very kind person from a large(ish) US tech company offered to help, and he and his colleagues provided enormous amounts of support in helping me produce a thesis which used real-world data to tackle a commercially-relevant problem. The experience taught me a lot about how Data Science actually works in large organisations and I was able to talk about what I'd learned (and contributed) in subsequent interviews.
My Advice to you would be the same: don't pour your energy into a niche academic project which has no relevance to your future ambitions. Find a problem faced by a real-world company and pitch them a solution to that problem using Data Science. If you can convince them that you might be able to help, they may be willing to provide data and a mentor to help you write a killer thesis and kill two birds with one stone.
Don't pour your energy into a niche academic project which has no relevance to your future ambitions
If you can't write your thesis with a real-life organisation, at least find a real-life dataset and work on a problem that ‘s relevant to the kinds of companies you want to work for. If you're stuck for ideas, you might want to check out this article where I talk through my own framework for coming up with Data Science project ideas:
How to Find Unique Data Science Project Ideas That Make Your Portfolio Stand Out
Don't apply to any jobs for at least the first 6 months
If your master's degree is one year long, I don't think it makes sense to apply to jobs for at least the first 6 months.
Why? Firstly, because applying to lots of jobs is a surefire way to suck the joy out of what you're doing in the moment. If you're only going to be studying for a year, don't squander that time: make the most of getting stuck into your course and meeting new people. Enjoy it, and don't spend all your time thinking about "what's next." Tomorrow will worry about itself.
The second reason you should delay making job applications is because you're unlikely to have much success at the start of your degree. Sure, you may well get through the first stages and be invited to interviews based on the strength of your academic credentials. But if you haven't actually learned much Data Science yet, then you won't have the skills to get through the technical interview stages. It's much better to wait until you've built your knowledge and skills and have something more concrete to offer prospective employers.
If you're thinking, "I can't wait 6 months to apply – I need to get something sorted NOW!" then I really get where you're coming from. When you're in your final year of university, it can feel like there's a lot of pressure to apply for jobs early on, especially given that lots of large companies typically recruit for roles near the start of the academic year. If that's you, my advice would be: don't let the pressure get to you. Apply when you're ready. You will definitely find something, and if you really want to apply for a milkround job then just apply next year and take a gap year while you're waiting.
Don't pick modules based on how likely you are to get a good grade
During the first term of my master's degree, all students took the same courses and we weren't given any choice about which modules to take. Zero. Nada. صفر.
Personally, I liked this. It meant that we spent more time as a cohort and built a broad foundation in many different aspects of Data Science, which was helpful for me as someone who didn't know which areas I wanted to specialise in yet.
Around Christmastime 2021, however, I had an important choice to make: which optional modules would I take during my second term?
My first thought was to select courses based on their perceived difficulty. "After all," I reasoned to myself, "my aim in this degree is to get a good grade so that I'm more likely to get a good job afterwards."
The thing is, however, this is really dangerous reasoning. If you only ever choose courses/jobs that you feel 100% comfortable doing, you'll never really grow or improve your skills. You'll get stuck in the same rut, exercising the same skills over and over again. Sure, you might do pretty well grades-wise. But you won't grow.
Based on my experiences, my advice would be to flip the tables on this way of thinking and pick the modules that scare you. Why? Because university is a very unique opportunity for personal growth. Compared to working full-time in a corporate job, you've got a lot more freedom in terms of how you use your time. In my experience, you won't grow much if you use this time to stay in your comfort zone. Instead, try some new and scary-looking courses and see what happens. Chances are, you'll find out an awful lot about what you enjoy (and what you don't enjoy), which will help you make a much more informed decision about your future.
In my case, this worked out pretty well. I took a scary-looking course on Natural Language Processing (NLP) – a course which I knew I would struggle with – but ended up getting much more out of it than any other course I'd taken at Oxford. It catapulted me into a NLP Data Scientist job immediately after I finished the course, and sparked a sustained interest in the field. I never would have known, if I'd not taken a chance on this scary-looking course in the first place.
Showcase your work as you go along
While I think it's a good idea to wait for a while before applying to jobs, it's definitely NOT a good idea to wait too long before starting to showcase your work and put together a portfolio.
When the time comes to apply to jobs, you will want to spend your time tailoring your applications and prepping for technical interviews. You will not want to be trying to pull together a slap-dash portfolio in a rush.
To avoid ending up in that situation, try and showcase your work as you go along. At the end of each assignment on your degree, after you've submitted your work, make a GitHub repo to store your code and a write a brief project description in the README.md file. Then, when the time comes to apply to jobs, you'll already have a repository of projects you can showcase to employers and use to prove that you are a fantastic Data Scientist. If you want to take this a step further, you can use a no-code platform like datascienceportfol.io to build a nice visual frontend to your GitHub repo. In case you're interested, here's the portfolio I built on the back of my master's degree:
If you've read any of my previous posts, you might be getting a bit bored of me banging away at this same drum ("make a portfolio!"). But, in my experience, you're doing yourself a huge favour in the long run if you take a few minutes to collate and showcase your work.
Not sure if a degree is right for you?
The final thing I'd like to say is that while a degree is an amazing way to grow yourself personally and professionally, it's not right for everyone. In the article you're reading right now, I've tried to give some advice on how to get the most out of a degree once you've decided to enrol. BUT, if you're on the fence, you might want to check out this piece I wrote on the things you need to consider before deciding to do a master's degree:
8 Things You Must Consider Before Committing to a Data Science Master's Degree
If you'd like to get unlimited access to all of my stories (and the rest of Medium.com), you can sign up via my referral link for $5 per month. It adds no extra cost to you vs. signing up via the general signup page, and helps to support my writing as I get a small commission. If you can't afford this (I'd completely understand!), it would mean a lot if you followed me. Thanks for reading!
