The Dataquest Download
Level up your data and AI skills, one newsletter at a time.
Don't Look at Job Titles — Do This Instead
|
I’m excited to welcome 2,707 new learners this week! In this edition, we look at how to overcome “research mode” when learning programming, we reveal our new “Classification with Neural Networks” lesson, and we highlight some standout projects from our community. As always, we’d love to hear your feedback. If you have questions, or if there’s a topic you’d like us to cover, just reply to this newsletter. |
Over the years, I’ve talked to thousands of people learning data and programming skills. Most successfully start new careers, but some don’t. The biggest difference I see between the two groups is how they approach learning.
When you’re learning programming, there’s an overwhelming amount of information to take in. Successful learners recognize this and focus on what’s important. This lets them get to action mode — learning and writing code — as quickly as possible. Unsuccessful learners can’t parse all of the information and get stuck in research mode.
The first opportunity to get stuck in research mode is before you even really start learning. It’s when you decide what you want to learn.
Most people who learn data and programming skills do it to get a new job. They need to research what job is right for them. Some amount of research is helpful. But until you understand some programming, it’s hard to understand what differentiates a database architect from a data analyst. This can lead to a frustrating rabbit hole of comparing roles without a good mental model.
If you’re finding that you’re getting stuck in research mode, there’s a mindset shift you can use to help. Most programming jobs require very similar skills, like problem-solving or system architecture. As you learn these skills, you’ll find that you can make a better choice about which career is right for you.
You can make a skills map by first finding some jobs that you’re interested in. Write down the top three skills you need for those jobs. Odds are high that you’ll find some commonalities across the jobs. If you don’t, go to a higher level of abstraction. For example, a granular skill is k-means clustering, and a more abstract skill is machine learning. The more abstract the skill is, the more likely it is to be common across multiple jobs.
Once you have some skills that are common to all of your jobs, start learning them! After learning a few skills, come back to the list of jobs. Now that you have more knowledge, you can evaluate which job is right for you more effectively. Then you can add or remove from your list of jobs.
This way, you won’t get stuck in research mode. You’ll learn and do research simultaneously.
Community Spotlight
The first iteration of the Community No-Lose Lottery is finished, and we’ll announce the results on April 10. In the meantime, a new iteration of the lottery began on April 1 and will last until May 31 — come join in!
This week, we have three impressive Community Champions — all of whom have completed one of our guided projects . . .
@tosingeorge01 shared her project on Hacker News Posts that is absolutely fantastic: comprehensive code comments, exemplary code style, easy-to-follow narrative, and interactive Plotly plots. In addition to the guided steps, Tosin also used linear regression to predict the number of comments and posts!
@FilipeFava also went beyond the guidance and learned new techniques to create an advanced project on Helicopter Prison Escapes. His work is a great example of the effective combination of storytelling and data visualization — and an excellent reference point for anyone at the beginning of their data science journey.
@giorgia shared an impressive project on Building a Probability App for Lotto 6/49 Players to treat gambling addiction, wherein she showed off her technical skills by providing an in-depth description for each created function. She also showcased her storytelling skills by briefly discussing the chances of winning the lottery.
Product Updates: Classification with Neural Networks
We’re back with the launch of our next Zero to GPT lesson, “Classification with Neural Networks.” In this lesson, you’ll learn that neural networks are complex models that try to mimic how the human brain processes classification, and you’ll
apply your newly acquired skills to predict whether a telescope saw a star, galaxy, or quasar!
We’ve also been working on a project for Introduction to Python for Web Development. In this project, learners will recreate the popular game Wordle using Python.
And last but not least, in the constant effort to offer the best data science learning platform available, we’re always improving things. This week, we added GPU support to our Guided Project offering to support Deep Learning projects — look for it to roll out in the next few weeks!
Keep Learning: Your Goals Are Within Reach
“Dataquest is changing the way we do our work fundamentally. We’re trying to move most of our team away from simple data reporting and business intelligence to more advanced analytics and more data-driven insights. Dataquest is making that transition easy.” —David Damberger, Chief Data Officer, M-Copa
|
See you next week! —Vik and the rest of the Dataquest Team |