What’s in a 9-weeks Data Science bootcamp?

Picture this for a moment. One year ago, before the pandemic hit us at full speed, I had recently onboarded a new client, prospects were bright and I had no reason to worry about the future of my consultancy. Fast forward to December 2020, my business is in a complete shambles, and clients are cutting costs left-and-right, even though the gaming industry has never been stronger. I had to pivot.

And today? I can’t believe that it’s been 3 months since I finished the Data Science bootcamp from Le Wagon, fresh-faced and rested after the holiday break. Everyone I asked about the bootcamp warned me: « it will be intense, you’ll be exhausted, and sleep won’t even come as solace since you’ll be dreaming about code ». I remember I filed this in my mental cabinet, not regarding it as the most useful advice I should heed about the next 9 weeks of my life.

A few months later, it’s time to reflect on this whole adventure. Am I an expert Data Scientist yet? Is it the hardest I’ve worked in the last few years? And more importantly, how often does Kaa manage to slither into my dreams?

#1: It IS intense

Since I’ve been a freelance consultant for the last 5 years, I’ve got used to setting my own schedule. While I’ve had crunch periods, they have seldom lasted longer than a week. My usual week / month was more of a healthy mix of hard work and pretty laid-back times, spent training myself and connecting with people.

For the 9 weeks of the bootcamp, I’ve worked. And worked. With the occasional yoga session to give my body a rest from sitting that long in my ergo chair. But mostly, I’ve worked. And slept. See #4.

A typical day starts at 9AM with a 90-120min lecture (some topics harder to comprehend than others, see #2), then we have until 5PM to put all this new knowledge into application by finishing up 3 to 5 challenges. As the challenges’ difficulty gradually increases, it’s not rare that I can’t finish them all when the 5pm bell rings. The day then ends with a last ‘recap’ lecture, more of a live coding session using all the concepts seen during the day.

Some days I had strength left in me to practice my flashcards or finish the challenges, but my usual evening routine consisted of a 30-minutes nap (fusing with my couch is more like it 😴).

#2: You gotta choose your battles

This is a tip I read halfway through the bootcamp on Frederik Durant’s blog.

Had I known beforehand, I would have avoided a guilt trip when I tripped on the most basic Algebra and Statistics concepts (derivatives, f-statistics and all these dark memories from high school 💀).

Having a CS background, 🐍 Python and SQL were easy. Even Machine Learning and Deep Learning, if I’m not trying to understand how the models work under the hood (I’ll dive deeper into this in the future) weren’t that hard to comprehend and apply.

Understanding how to tune and optimize your model is another story, as most of these parameters are obscure and their impact hard to understand. I mean, how are you supposed to know what value to put as the alpha or L1 of the ElasticNet model?¹

Since our teachers are always quick to reassure us that it comes with practice and experience, I’m not worried, it’ll all come to me at some point.

#3: Studying remote is a better experience than I expected

The main reason I decided to attend Le Wagon in Brussels instead of Lille or Paris is the proximity to the European Commission, as I aim to put my newfound knowledge to good use, using all the data produced by the various EU member states to inform on new policies.

As COVID is still a strong part of our life as of the time this writing, I’ve been unable to travel to Belgium and attend the lectures on-site. Approaching the start date of the bootcamp, I was starting to worry about what studying remotely would be like.

Well, I’m glad to say that it’s been a breeze. Anyone who knows me won’t be surprised to hear that I missed the interpersonal connections that are built when you’re sharing a common space, but Zoom proved to be a great tool for class attendance. The tools built by Le Wagon to ensure you’re not working everyday with the same person are working like a charm.

Having the bootcamp 100% remote also allowed for non-Belgian students to attend: Batch #557 was a great multicultural group of Belgian, French, Austrian and German students.

#4: Do Data Scientists dream of electric snakes?

Data Science students sure do.

I couldn’t count the times I woke up in the middle of the night thinking “hah, that’s how I should have done it” 🤯

#5 Don’t stop learning after the bootcamp

That was the big question when I started out.

Would I be a good Data Scientist after this bootcamp?

Well, no. The breadth of topics touched during the bootcamp is indeed super wide: going from Python 101 to Convolutional and Recurrent Neural Networks in just 7 weeks is no small feat.

The downside of this is, what we’re doing is brushing out the surface of everything. While this does not make you a full-fledged Data Scientist yet, it serves two purposes:

1. If you have an existing business professional or academic background, it’s the perfect stepping stone to starting out a new career in Data Analysis. You’d be operational and autonomous right away, able to use your newly acquired skills in any industry.
2. If you feel like pushing your education further, it’s easy to do so via either another bootcamp, a university degree or working on side projects to teach yourself the required skills. What’s more, you already know what your strengths and weaknesses are, so you know where you can shine easily (to build that coveted Github profile) and where you would need to dive deeper into the concepts and applications.

One thing that I wish we’d cover more in details is the ethical and environment implications for Machine Learning and Artificial Intelligence. We had the opportunity to attend a private screening of Coded Bias with its director Shalini Kantayya, but apart from this, not many opportunities to discuss it with teachers and speakers.

All in all, I’m very satisfied with the bootcamp. It’s been an eye-opening experience and a very good first foray into the Data Science world. I now know what I want to work on in the future, and most importantly where I should focus my learning efforts in the next few weeks before landing a job where I’ll grow, surrounded by a great team of mentors and coachs.

Photo by Lucas Law on Unsplash