Garage

My experience mentoring Tanmay Bakshi at the Toronto IBM Garage

Share this post:

For three months, the IBM Canada Lab had the opportunity of hosting Tanmay Bakshi, a software programming wunderkind from Brampton, Ontario, through a unique student training program tailored just for him. At 12 years old, Tanmay is taking the developer community by storm with several iOS apps, his own “Tanmay Teaches” YouTube Channel, and public-speaking engagements where he has shared his expertise in front of thousands. Throughout the three-month program, Tanmay spent time at the IBM Canada Lab in Markham and the IBM Bluemix Garage in Toronto, learning from seasoned IBM developers, where he enhanced his skills using IBM Bluemix and IBM Watson, and sharing the unique perspective only a 12-year-old whiz kid can impart. Since June, Tanmay was paired with IBMers who helped accelerate Tanmay’s growth as a developer and thinker.

Here is the perspective of one of his mentors, Ankita, who is a developer at the IBM Bluemix Garage in Toronto.

I am a developer at the IBM Bluemix Garage in Toronto, a consultancy with the mindset of startups where we embrace Extreme Programming, IBM Design Thinking, and Agile methodologies to help clients build innovative, digital solutions. I love being part of the Garage because of the variety of projects that come in, which often allow me to test early technologies. But it’s not only challenging projects and new technologies that make the Garage exciting, the people at the Garage are fun to work with.

20160819_140154

Over the past few weeks, I have been mentoring Tanmay Bakshi, a 12 year old developer who creates cool YouTube videos and shares his knowledge with his amazing public speaking skills. With the help of John Marini, the Practice Leader at the Toronto Garage, and IBM Design Thinking, Tanmay came up with the idea of building an application that provides secure access to authorized personnel and prevents false identification. This app is unique in the cybersecurity space with its use of the Watson Visual Recognition API and its development using IBM Bluemix.

The IBM Bluemix Garage Method

We started this project by taking Tanmay through an IBM Design Thinking session – a hallmark of the IBM Bluemix Garage Method. Using IBM Design Thinking, we constructed an empathy map to put ourselves in the shoes of our end user, and to learn more about what he or she is doing, thinking, and saying when verifying their identity. We examined the existing situation, identified places for innovation, and selected the “wow” factors in our app. Of course, to keep ourselves in check, we defined goals to measure our success. We were able to define our Minimum Viable Product (MVP) which is to provide secure access with confidence, without having to carry a badge or an ID. Our end user was an employee of a company, who carries a badge or other form of identification, but hates carrying it and also has a fear that he or she will get locked out. Our MVP helps us provide secure access with confidence, without having the employee carry a badge or an ID. After brainstorming and looking through the library of APIs available on IBM Bluemix, we found that the Watson Visual Recognition API could help us identify people and prevent a security breach.

20160819_14011020160819_14011320160819_140117

We also identified the MVP by determining what was achievable with the time and resources available. I started mentoring Tanmay when he came to the Toronto Garage every Friday for shadowing and training in the Garage, and a few additional hours in the evenings for pairing. Although Tanmay was experienced in building mobile apps, he wasn’t familiar with web development. This was great as I was able to demonstrate the IBM Bluemix Garage method through that process. In the Garage, we believe in pair programming as part of our culture so that anyone new to any technology can pick up on things pretty quickly. I personally had less experience with Raspberry Pi, so Tanmay and I learned that together. We determined risks in our project every week and worked toward our objectives. We kept the emphasis on our MVP and started setting achievable goals.

Why Watson?

Watson is one of the greatest Artificial Intelligence (AI) machines out there. Like most AIs, the Watson Visual Recognition requires training just like any other AI would. It requires positive and negative samples. In our case, positive samples represented people who we (Tanmay and I) want to provide access to and negative samples represented people we want to deny. Watson needed 50 pics each of positive and negative samples. Watson was the simplest part, as we just had to perform 3 REST calls to get our data trained. Watson spits out confidence level on a random image provided by the developer.

 

The End Product

20160729_132903

We picked Raspberry Pi3 to build the IoT (Internet of Things) part of the solution using Python libraries. As you enter the room, a motion sensor detects motion and you’ll be asked to identify yourself. Raspberry Pi3 will take your picture using the tiny camera attached to the Pi itself and upload it to the Cloud where the Watson Visual Recognition API will take care of the rest.

Watson will then perform an enhanced analysis and accordingly provide 2-factor authentication. If the confidence level provided by Watson is greater than 40%, it will grant access; otherwise, it will prompt you to enter your email and send a passcode to your mobile device using Twilio.

My experience with Tanmay

Tanmay picked up new technologies at the Garage very quickly and asked a lot of great questions, which helped him fit in easily. But, perhaps what is most unique about mentoring Tanmay is that when we were talking with him, we didn’t feel like we were talking with a 12-year-old; he felt like a peer.

Tanmay is taking home the effectiveness of pair-programming, and most importantly the whole IBM Bluemix Garage Experience with greater technical skills such as how to build and deploy an app on IBM Bluemix using AngularJS, Node.js, Watson, IoT, Bootstrapping, and also the small tricks that we picked up from each other while pairing, to make our lives easier as developers.

I wish Tanmay good luck and hope that his time here at the Garage was very valuable – but most importantly fun!

Resources

Add Comment
No Comments

Leave a Reply

Your email address will not be published.Required fields are marked *

More Garage Stories

Watson is getting a little emotional with PubNub

For my first API World session, I'm sharing the stage with Bhavana Srinivas of PubNub. Together, we'll cover how to make cognitive, real-time apps using IBM Watson services with PubNub Functions. As you will see at our session, combining Watson Tone Analyzer with PubNub is not only lightening fast, but it's also very easy.

Continue reading

API Connect can become your tool of choice

At this year's API World, I'll be hosting a workshop with my colleague, Marek Sadowski. Some of you will know Marek from his robot-building exercises. Our workshop will focus the power of API Connect bringing together disparate worlds for the benefit of developers.

Continue reading

Top 5 Highlights from VMworld 2017

With VMworld US & Europe coming to a close, we’re taking a look back into activities and announcements that took place over the past couple of weeks. Below is a recap of some of the most exciting stories for those who didn’t get a chance to hear the news on the ground.

Continue reading