Sharing the readily available online resources that allow for self-learning
Like everyone else, Google is my best friend. Pratham Prasoon has shown me how the internet has liberated learning for everyone. Machine learning is no longer a complex subject limited to academic folks. Young and old alike, can have access to the common resources and are free to learn what they like.
Read on to find out how Pratham picked up his skill sets and what it’s like to be a young person in data science in current times.
Huifang: Where does your love for technology stem from? And, what motivates you to continue pursuing it?
Pratham: I’m not quite sure why, but I was always on the computer playing some old games since I was very young. I was just curious about how these things generally work, so I kept exploring.
This exploration took me into machine learning a couple of years ago. However, there weren’t many good resources in the bookstores for me to learn the topic. So it was put on hold until this year when I had some free time and acquired the necessary resources to learn it.
So, it all boiled down to my curiosity, and this made me the techwiz I am today.
Huifang: What was the first programming language you picked up? How did you do it?
Pratham: The first programming languages I picked up were QBasic (which is a shortened name of “Quick Beginners All-purpose Symbolic Instruction Code”) and Python in early 2015.
(Huifang: He was only 10 when he first started.)
My mother put me in a two-day workshop where I had an introduction to the programming basics such as printing statements, loops and Mathematical functions. Then, I kept becoming more curious about what I could do with programming. This led me deeper down the rabbit hole.
Huifang: Tell us more about your learning experience. Where do you get your learning materials?
Pratham: I didn’t have good internet access until four years ago, so I didn’t have anywhere to learn programming from. I stopped programming for a couple of years and started back on it a while ago.
I mostly self-learned from articles and tutorials. freeCodeCamp is the first place I go to for my learning. And there are also lots of YouTube channels with really good content.
Some other resources include:
- Traversy media (Web development and Programming tutorials)
- Sentdex (machine learning, finance, data analysis, robotics, web development, game development and more)
- Techwithtim (programming, software engineering, machine learning and everything tech)
You can always Google search and find good and relevant articles from Medium.
I mostly learnt programming by Googling issues that I came across during coding and trying to understand why the issue happened.
Honestly, I didn’t have anyone to help me when Google search didn’t work. So I would just try random solutions until something worked. It was really a struggle for me. For instance, one of the most difficult things to understand when I first started out was using the terminal and running commands. Google search, in this case, couldn’t address all the issues that I’d faced.
Huifang: How do you think technology has evolved the way that people learn?
Pratham: Before the age of the internet, buying books or going to college was probably your best bet to learn about computers. However, because of the widespread accessibility of the internet, people are sharing their knowledge through these online platforms and blogs and whatnot.
It has become much easier to learn things. But on the other hand, there is an information overload. There is too much information and technical jargon out there.
Huifang: You’ve grown your followers rapidly on Twitter over the last year. Why do you think that people are so drawn to you? Can you share with us more about your Twitter journey?
Pratham: I only started using Twitter actively when lockdown started. I immediately noticed a really good tech community on Twitter. After interacting with people and made new friends from the community, others also noticed my account and started following me.
At first, I used to tweet about web development. Later on, I started talking about machine learning. There are a lot of complex maths and jargon in machine learning. So I try to be as helpful as possible by sharing resources and guiding people in the right direction.
Although I don’t consider myself to be the best programmer, I try to share the knowledge I have in the best way possible. I think that’s the main reason I’ve grown.
Huifang: Now that you’re pretty well known and you got a job, do you get people who judge you based on your age and brush you off automatically?
Pratham: A lot. Ageism is a real thing.
I do not deny that I may be wrong at times. But sometimes people dismiss my statements even when I feel I’m right, just because I’m young.
I’ve learned to live with it.
Some people tell me that I should be spending more time on my studies and stop the “Twitter nonsense”. There will always be people who criticize me but it doesn’t mean I should stop what I’m doing.
I’m doing it not to prove anything to anyone, but simply because I like it.
Huifang: I interviewed Mabu Manaileng a while back and he mentioned that he wishes he’d started data science 10 years earlier. Since you started out so young, do you think you have an advantage over the others?
The previous generation of data scientists didn’t have the tools that I have right now.
There are lots of open-source frameworks that make data science very simple. With the vast amount of information available online, learning has become much easier.
Data science is becoming more accessible to people and less academic, in the sense that there are fewer academic entry requirements to it. I think getting a job in data science is still tough. However, for those without a degree in data science, such as those who are self-learned, it’s now relatively easier than before.
Huifang: Do you think one has to have a very good understanding of math before they can do machine learning?
Pratham: I passed on the idea of getting into machine learning four years ago because I thought the math behind it was very difficult. When I started machine learning last year, I realized that people overstate the amount of math required for it.
I’m not saying that math isn’t important, but you can totally do without it if you just want to start machine learning.
Perhaps machine learning used to be a very academic field that involves a lot of math. In recent years, there are new frameworks in place and tools that abstract the math for machine learning. So, in my opinion, all you have to do is learn programming to get started
Huifang: I got to know that you’re currently interested in learning about Generative Adversarial Networks (GANs). What interests you about GANS?
Pratham: As they say, competition makes you better.
Simply put, in GANs, we “put two neural networks to fight against each other”. In this GAN case that I was testing out, one part of the model created either a fake or real image. The other part of the model evaluated if the image created was fake or not. So through this process of training, the model gets to know what is real and what is fake.
This can be used to generate a lot of things, such as marking art. It’s cool, but there’s certainly an element of complexity.
Huifang: Who in the data science community inspires you?
Another person I followed on Twitter is Santiago. He makes great machine learning content which inspires me to go more into machine learning.
Huifang: What kind of content can we expect from you next?
Pratham: It’s going to be mostly on machine learning and programming.
I want to try applying my machine learning skills to financial topics such as personal finance and stock analysis. So I’ll try to incorporate some of these into my content.
There’s no reason for us to put anyone down because of their age. Nick D’Aloisio was only 17 when he designed an app worth $30 million. Jordan Maron, also known as Captain Sparklez, created his first YouTube channel when he’s 18 and he’s worth more than $8 million now.
Everyone starts somewhere and it’s great that Pratham started out young. I’m excited to see him grow with endless potential. If you haven’t read the first part of this interview, follow us on how age is only a number when it comes to data science.