Meet Tanmay Bakshi: an exclusive interview
Rushika Bhatia

Meet Tanmay Bakshi: an exclusive interview

Not every 14-year old wakes up in the morning looking forward to the convergence of deep learning models. But then, not every 14-year old is Tanmay Bakshi. Such a delight to feature this young genius…

Tanmay, you are the world’s youngest Watson programmer. Can you walk us through the journey of how this happened? What were the triggers that led to your interest in coding?

I originally started coding with the help of my dad, when I was five years old. After that, I started to use books and the internet as learning resources to help me continue my journey in the field of computing. When I was 8 years old, I learnt iOS programming, and at 9, had my first iOS app on the store. After this, I developed numerous iOS apps; but when I was 11 years old, I stumbled upon IBM Watson. From there, I had a new mission: to use AI and cognitive for the benefit of all humanity.

What are some of the apps that you’ve developed? Which one has been your favourite and why?

I’ve developed numerous applications, most of which use the power of artificial intelligence. My very first AI application was AskTanmay, an NLQA (Natural Language Question Answering) system, that can answer your person, organisation and location questions. I also recently developed FaceMLvault, which aims to use IBM Watson and deep learning algorithms to perform facial recognition. However, the one that I’m most excited about is called “The Cognitive Story” – it’s because this project is in the field of healthcare, the field that I think is going to be impacted the most through the power of AI. Through this project, we’re trying to bring artificial communication ability to those who can’t communicate naturally, by understanding their EEG (Electroencephalogram) brainwaves. I’m working on the deep learning aspect of this project, which powers the real understanding of EEGs.

Tell us about your work with Watson.

I’m passionate about services like the ones provided by IBM Watson due to the fact that they solve one of the main problems with AI: accessibility. Developers can incorporate AI in their applications, but because it’s so difficult they mostly avoid it. However, services like Watson make AI easy to use for developers – they just need to incorporate different types of tools! They just need to incorporate different types of tools in their applications. With Watson, I’ve been able to build lots of different applications. In fact, “The Cognitive Story” app I mentioned previously is also powered by IBM Watson – so are AskTanmay and FaceMLvault. I’ve worked with every Watson service available publicly, and have created AI-based solutions in practically every major field that AI impacts.

There’s a lot of talk about technology and its impact on our lives. How do you see technology shaping our personal and professional lives over the next five years? Do you think there will be a significant impact?

Whether you realise it or not, your life is already being shaped by artificial intelligence and technology in general. Whether you’re typing on your phone’s keyboard, taking a depth-effect picture, using Tesla Autopilot, or even have a Gmail inbox, you’re using and bene ting from the power of AI. However, deep learning is relatively new – not the concepts (those have been around for decades!), but the popularity and widespread use is new; it’s only around five years old. However, imagine, in just five years, we’ve slated greater than exponential growth, so I can only imagine that, once we’ve removed the public fear of AI, and put AI into the hands of more developers, we’re going to see AI take an even bigger part of our lives; it’ll allow us to do things we could never have dreamt of before.

Particularly with AI, there’s been an ongoing debate about its use and its coexistence with human intelligence. What’s your take on this?

My take is artificial intelligence has a lot of fear around it. This fear is caused by the fact that people aren’t in control of, or in the backend of AI. They aren’t able to see the way it actually works, they have this stereotype of AI from movies and TV shows. However, when you take a deep dive into the backend of such technology, then you start to realise that no, AI won’t just randomly turn against us and outsmart humans! It can’t – it’s not capable of doing so. Its objective function, or the function that tells AI what to do, its intent, is constant. While humans, themselves, may set this intent to something negative, we’ve been doing this for all technologies that have ever existed! Yet, we’re not stopping tech progress, we’re combatting against evil, AND we’re accelerating tech progress. Why would we stop for AI?

As cognitive services emerge to the forefront, what industries do you think will be affected the most?

I think all industries will be affected – reason being that ALL industries are being forced to be tech-centric ones. And wherever tech lies, AI will impact that spot. However, the more tech-centric and data-driven your industry is, the more AI will augment it. For example, Healthcare, Security and Finance are great examples of industries that are ripe for AI to augment.

What can businesses do to prepare themselves for this change?

Mostly, I’d say you need to be willing to accept the change and adapt it to your business as efficiently as you can. Plus, you shouldn’t have the wrong expectations: for example, there’s one expectation that a lot of businesspeople have from AI that simply isn’t true; it’ll replace their employees, and bring cost savings! It won’t replace them, it’ll augment them, make them more efficient, and allow you to save cost that way!

Terms like AI and blockchain have almost become buzzwords today. How does one take away the hype from what really matters?

While these technologies have a lot of hype around them, I believe it’s well deserved. However, the way AI and blockchain are sometimes portrayed in this hype is completely false – it sets false expectations, and when those expectations aren’t matched, people think that these technologies are useless and that we’ve just been hyping them up for no real reason.

How can communities use such technologies to make a social impact? Is there a mission close to your heart that you are passionate about as well?

As I mentioned, I’m really excited about how AI can be implemented in fields like healthcare, where it can save, augment and generally improve people’s lives. People, generally, can make social impacts via
this technology in really simple ways. In fact, open source technology is one the key factors in deciding how much we’re able to advance state-of-the-art AI algorithms. So, just doing something as simple as making sure your code is open source, and so are your algorithms and techniques, can make a huge difference. Imagine, if deep learning remained proprietary to everyone who developed it, we wouldn’t have ANY of the advances that we do today. Also, due to this, I have a goal to reach out to and help at least 100,000 aspiring beginners, to help them learn and innovate along their journey of programming. I’m already around 5,000 people there.

You travel a lot to keynote prominent conferences. What appeals to you the most about these large-scale events?

Mostly, I love the fact that I’m able to take such powerful technologies, and spread the word and my message about them, to so many people. AI is powerful – but we need people to open their eyes to the power that deep learning provides, and to how their lives are already surrounded by it! You’ve obviously come so far at a very young age. What’s next for you? Where do you see yourself in the next couple of years? Well, mainly, I want to continue researching in the field of AI: architecting better neural networks, and then applying those great architectures in the fields that need them the most, such as, Healthcare. I also wish to keep sharing my knowledge of these technologies, and not just stop at 100,000 in my goal but expand from there.