Who’s the best Data Scientist in the world? Cassie Kozyrkov? Yann LeCun? Someone else? Whoever this person is, you can be sure of one thing: This person has worked hard. Because to reach the top, it doesn’t only take talent but hard work. Hard work starts with motivation until persistence takes over. Indeed, motivation always wears off over time. And what do you do when it happens?
This is at this precise moment that you’re taking an important decision; should I continue or should I stop? Continuing is hard because it is hurtful, and stopping would be the equivalent of giving up. Giving up to who? To yourself. Because you, and only you, is your main competitor.
This article will share 5 tips to become better than your yesterday-you.
The 1% rule consists of practicing at least a little bit, but everyday. Basically, doing even a tiny little bit everyday proves in the long run to be more powerful than being irregular. Someday you’ll do great; someday you’ll struggle, but everything will eventually pay off.
The key takeaway here is to build a habit, a routine.
Put differently, we can also talk about memorization. We’ve certainly all spent late nights, the days before our exams trying to fill our brains as much as we could. Chances are that we’ve already forgotten most of this knowledge. The reason is simply because we’ve different parts of our brain who retain information differently.
Might come off as common sense; but the more time you give yourself to learn something, the more you will retain. Which can echo what we’ve just developed previously. Something you’ll learn on day 100 will resonate differently than something you learn on day 1.
Sometimes, it doesn’t really matter if you learn again something that you already knew. Since in between you’ve improved your understanding of the domain, your comprehension of it will have evolved too. And you will remember things that you didn’t last time, or understand things differently.
On the back of what has been just said, going through the basics over and over can be powerful. For instance, someone mastering the ins-and-outs of Pandas can bring a lot of value. Potentially more value than someone playing around with cutting-edge technology.
Pandas and SQL are the bedrock of any analytics project. A deep understanding of these 2 tools enable a great optimization of your work. Someone knowledgeable will know what data type or what function to use that runs 10x to 100x faster and ultimately contributes to ML ops success.
Equally, being unbeatable with the basics will make you faster at delivering results. In industry, being able to deliver proof of concepts, 1st iteration of work, or quick & dirty ad-hoc solution satisfy most of the stakeholders more than we think.
The Lindy effect is a theorized phenomenon by which the future life expectancy of some non-perishable things, like a technology or an idea, is proportional to their current age. — Wikipedia
Not only will it optimize your projects, and make you more efficient at your job, but it will also make your role more future-proof.
The basics, by definition, have been in the game for a long time. And statistically, the longer they’ve been in, the longer they are likely to remain. With this in mind, you’ll protect your career significantly if you decide to become the master of something standing the test of time rather than trying to pick all the new trades.
Don’t get me wrong. I’m not advising you to not give new tools a go, but use your learning-time strategically instead. 80% on the basics, and 20% on the new for instance?
One needs to work strategically. Your learning time is limited; therefore you need to focus on what’s really important.
Kaggle comes in handy for that matter. It allows you to look at the best, and identify what separates you from them. Alternatively (or additionally), you can attend webinars or local meet-ups. Identify things you don’t know. Then, you can build your learning plan accordingly. Ultimately, it will help define an “As it is” versus a “To be”, which will naturally design your learning plan. The key here is to be curious, to google everything that you don’t know or are unsure about.
The Dunning–Kruger effect is a cognitive bias whereby people with low ability, expertise, or experience regarding a certain type of a task or area of knowledge tend to overestimate their ability or knowledge — Wikipedia
To further expand on the previous point, this knowledge map relates to our own bias. Hence the importance of putting effort into mapping out what we don’t know. Only then will one realize where they really stand on their learning journey.
Your strength is only equal to your weakness. On the importance of knowing yourself. If you’re great at fine-tuning models, story-telling, data visualization but it actually struggle to clean a dataset this will catch you up.
Fortunately, today resources to learn are abundant. Online courses, Youtube videos, newsletters, podcasts. Name your favorite way of learning, and chances are that a support for it exists. Be specific in what you learn. And work on your main weakness.
Subscribe to newsletters with your professional email address and and read at least one new thing per day. Another idea, start the day with a routine of reading articles, or doing a training exercise as per point #1 of this article. You can also create a list on twitter of data analytics experts.
When you learn a new language, you’ll hear plenty of new words. It’s hard to retain them all. So a good way to remind these words is to use them as soon as possible after hearing it for the first time.
Reading and listening are passive, speaking and writing are active. Learning actively has proven more powerful than simply learning passively.
It is the same for data analytics. As much as you can, try to repeat, share, teach what you’ve learnt. Whether it be a show-and-tell at work, giving a a public talk, or simply writing an online article anonymously, there’s no bad way of spreading your knowledge.
The bottom line
In this article I’ve tried to share five immediately actionable tips that will help you become a better analyst/scientist than yesterday. All in all, it doesn’t have to take you hours per day; it just needs to be constant. Also, it goes without saying that you should be kind to yourself. Looking up at other people or what you can be is a great motivation, but don’t be disappointed if you don’t meet your expectations. Set up a system, trust the process, and enjoy the ride.
If you enjoyed this piece, please make sure to follow my profile, so you don’t miss any of my weekly articles.