An Interview With Tanmay Bakshi, The World's Youngest IBM Watson Developer
Also a Keynote TEDx speaker, Author, YouTuber, and More!
When was the last time you came across a headline exploring the marvels of artificial intelligence? Today, the progress in this field is at an unprecedented rate influencing multiple industries ranging from healthcare, finance, and astronomy. If you thought this domain is only for the PhDs and other senior researchers, you are wrong.
He built his first iOS app at 9 and had published a book by 14, not to mention the numerous keynotes he has given in conferences across the globe. Having already built some impressive deep learning projects like AskTanmay, he is the youngest IBM Watson developer out there.
Tanmay Bakshi comes off as your average teenager invested in sports and watching TV, but he’s also the child prodigy whose achievements don’t end with what we’ve listed above.
How many middle-schoolers in your neighborhood do deep learning? For Tanmay, it was his passion for coding that brought him where he is today.
Read on as he shares his views and experiences with Probe, NIT Trichy in this insightful interview.
Q 1: At the age of nine, children would probably worry about how to skip school for the day, but you already had an iOS app published by then. How did it all begin?
Tanmay: Originally, I began to code when I was 5 years old when my dad introduced me to simple languages like FoxPro and Batch. From there, my curiosity lead me to learn more through resources like the internet and different books. I learnt languages like Visual Basic and Objective-C. Objective-C led me to learn iOS Application Development, and the first app of mine published on the Apple App Store was tTables — an application that helps you learn your times tables.
Q 2: Your childhood experience was probably different from what most people generally have. Does that make you feel that you missed out on anything? If you wished to do something differently, what would it be?
Tanmay: During my childhood, I was playing with computers just like any other toy of mine, which is why I got so much in love with coding. My curiosity of how to not only operate a computer but also to learn what goes behind its working, the code, is what made me passionate about tech!
I truly don’t feel like I’ve missed out on anything because I continue to do what I love to do. I don’t code, for example, because I have to, it’s because it’s been my passion for over a decade now! Therefore, I can’t imagine myself being passionate about anything more than coding. Apart from coding, I love to share what I learn, through my YouTube channel called Tanmay Teaches, the books I author, the keynotes and workshops that I do, and doing all this makes me realize the power of code that helps me connect with millions of people, and I always cherish that!
Q 3: You have a rather unconventional educational background, having been home-schooled for quite some time. What is your view on the traditional education system, especially the scenario in India? How much value do you think should be placed on college degrees and similar credentials?
Tanmay: I do believe that college degrees, and schools overall, are important.
I’ve always believed in self-learning because love for learning leads to actually mastering the subject. Practice and example based learning work the best. If you have a degree that you’ve earned by gaining just the theoretical knowledge, neither you nor your employer will benefit much from it. It really depends on an individual’s career path that decides if the degree or the skill or both should be given more importance.
The traditional education system and the governing bodies over them need to adapt to new technology — they need to evolve. Schools and universities are all about getting the next generation ready for the future. If, for instance, in a hospital, you’re using an outdated version of Windows, it’s not optimal of course — especially from a security sense. In a school, when you’re not using new technology, you’re not teaching the next generation to use that technology, they’re not going to be ready for the jobs that require them to use it to its full potential. Also, it’s not just teaching kids how to use technology that matters, that same technology will also help teachers teach their students in a much better way. You see, there’s a limit to how much humans can do — a teacher can’t answer every question that every student has, specially if they’re all meaningful, thoughtful questions. When teachers leverage technology, specially, the next-gen technology, like Machine Learning, they have the opportunity to teach their students in ways that are exponentially more efficient than what they use right now.
Q 4: If you were given access to all the resources on the planet, what would you work on? What kind of problem would you yearn to solve?
Tanmay: Machine Learning (ML), the technology I’m most passionate about, is a technology that requires a lot of resources. Usually, you hear most about the fact that you need a lot of computational power for machine learning to be effective. However, there’s one more thing that ML requires, that doesn’t usually get enough attention: data! Some of my projects are currently progressing slowly because there’s not enough data to train the machine learning algorithms behind them. For example, there’s a project of mine in the field of Mental Health, around predicting and preventing depression at a very early stage, to stop suicide at the very first step. The problem? There’s a lot of data out there, but due to restrictions, all that data needs to be anonymized. So, if I had all the resources in the world, I’d work on implementing machine learning in Healthcare: specifically, in fields like Mental Health.
Q 5: Your informative YouTube channel reflects a lot on how much you love to teach. What was your main inspiration behind starting to teach?
Tanmay: If you take a look at some of my first YouTube videos around different tricks in the macOS Terminal, and using Visual Basic for simple tasks, my inspiration was to create a kind of “repository” of things I’ve learnt how to do — and in the process, enable others to learn these things, and answer any questions they may have. Over time, though, I started to realize that there’s a very steep learning curve for people who are starting to get into technology, all because there’s a lack of resources. I’d be getting tens of questions, right as I upload a video, on how to start coding, implement certain algorithms, or fix certain errors in people’s code. I realized this and this then became my main inspiration. I initiated my goal to reach out to and help at least 100,000 aspiring coders to help them innovate along their journey of learning to code. So far, I’m around 11,500 people there.
Q 6: Your achievements at such a young age are absolutely amazing. With a promising future ahead of you, do you have any specific plans for your career? Any interest in entrepreneurship?
Tanmay: Thank you! In general, there are 3 things I want to do: first, I want to continue developing with technology, therefore, creating more content for aspiring innovators to learn from my YouTube tutorials, the books that I author, and the posts that I write. Second, I want to enable more developers to work with machine learning technology and I do this by sharing whatever I learn with the open-source community. Finally, I want to continue applying technologies such as Machine Learning in fields like healthcare, where it can save, and augment, the lives of millions. In terms of entrepreneurship, I’m not doing any of this with the intent of starting a company yet. There’s still some time for me to decide!
Q 7: With data becoming increasingly essential for the development of AI, What are your views on data privacy? Do you believe that companies should be allowed to store user data on such a large scale?
Tanmay: As I mentioned previously, data is a necessity for machine learning technology — and ML is something that impacts the user experience of technology at such a fundamental level, that people don’t even realize they’re using it. When you use Siri, or when Gmail predicts the rest of your sentence, you don’t actively realize “hey, machine learning is great!”. However, without it, humans would spend a lot more time doing things that could be done by our machines — enabling us to spend our time and energy on the things that really matter. So you have two seemingly conflicting, but equally as important, premises: you need machine learning, but you want to keep your data to yourself. Contrary to what you may believe at first thought, there are ways to get the best of both worlds. For example, what Apple does to train machine learning models for their QuickType keyboard is — they train models on-device and aggregate model parameters on their cloud. So, in short, I do believe that companies should be allowed to store user data at a large-scale, but only with explicit permission from the user, and in a scenario where users know exactly how they benefit from providing their data.
Q 8: How important was the support of your family in your journey so far?
Tanmay: My family — and the huge support they have been — has not only been important, but crucial. Without them, I wouldn’t be living my dream. I’m very fortunate, for not only did my family value what I’m passionate about, but they also trusted in me in taking that ahead. I’ve always had their unconditional support and guidance along the way.
Q 9: What is your advice to young aspiring students wanting to get into the field of AI?
Tanmay: Well, there are a few steps to learning about AI/ML. First, you’ve got to build a strong foundation — and that foundation is code. You can only implement machine learning if you understand how to code inside and out, and your life will be unimaginably easier if you learn to code in the languages that the machine learning community is already hovering around like Swift, Python, or Julia. Once you’ve learnt to code, you can get into the fundamentals of machine learning by reading papers on, well, fundamental topics like gradient descent/backpropagation, and more. Finally, I’d recommend you learn by example, by coding with libraries like NumPy, SciKit-Learn, Swift for TensorFlow, PyTorch, Keras, and more. This’ll help you be part of the mainstream.
Q 10: We have to ask you this one to satiate our curiosity. What else do you work on in your free time? Any specific hobbies? Do you follow pop culture, movies, or such?
Tanmay: Of course, my favorite sports are biking, table tennis, and cricket. I love watching the TV Show called SpongeBob SquarePants, and I could watch that all day — only if my mom would let me ;)
Tanmay Bakshi’s laurels don’t end here. His book, Hello Swift! IOS Programming for Kids and Other Beginners is a great place to start learning Swift, the programming language used to build iOS apps and he also teaches on his Youtube channel with over 300k subscribers, Tanmay Teaches.
You can view his work on GitHub here. To know more about Tanmay, check out his website here.
Probe, NIT Trichy is immensely grateful to Tanmay for sharing his thoughts with us and would like to wish him success in all future endeavors.
Editorial Note:
This interview article was compiled in collaboration with Nivetha Balu of the Probe content team.
A special mention to Karthik Swaroop and XQ from Probe for contributing questions to the interview.
Clap and share if you liked this one and do follow Probe, NIT Trichy for more interesting interviews.
