A few years ago, someone pointed out to me, “What you are pursuing is called data science.” Surely enough, the remark left me intrigued because up until that point I’d only seen myself as an engineer, a designer and occasionally, a machine enthusiast. During college, a huge chunk of my initial work involved robotics and product design; and veering into data science wasn’t even remotely on the cards. Needless to say, once I did start as a data scientist, a lot of my friends and acquaintances had questions about my choices. Now when I look back, there’s no one reason I can pinpoint at to explain what led me to where I am today, but a couple. To put it briefly, I believe that the projects I choose focus more on the impact they have on people, and not the engineering skills required. 

Hang on – I think it would be nice to steer clear of the ambiguity once and for all. So, I am jotting down my journey to highlight the major events that lead me to data science.

Introduction

I’ve been an avid explorer for as long as I can remember. Over the years came in a flurry of hobbies & passions, and while some faded away within weeks, a few remained. With science & maths being my long-lasting interests, as one would expect, pursuing engineering was an obvious choice. With some sweat and toil, I landed up in the Indian Institute of Technology Madras and enrolled for a 5-year program in the Department of Engineering Design. The program is a unique amalgam of courses ranging from applied engineering to design, as the name suggests. On a typical day at college, departments engage students until 5 pm. Sports, robotics and many other clubs take over after. Among many such clubs, the Centre of Innovation (popularly known as CFI) is another dimension to innovation. Giving shape and form to all engineering ideas, many students spend hours on end here.

Engineering | First Year, IIT Madras

CFI lured me in the freshie year itself. I constantly toyed with ideas all through my school years, but it was a presentation by Mr. Ashish Bajaj (Head of CFI), which proved to be the focal point of change. A vast hall filled with first-years watched the CFI introductory video which showcased previously designed robots by students, and it left me with goosebumps. I still remember a little voice in my head telling me to work towards getting featured in next year’s orientation video.

CFI is the birthplace of several innovative projects, but a wall-climbing robot had never been conceived before. However, the catch was that CFI only approved problem statements that solved a real-world problem. Endless hours of frantically googling finally gave results; the concept was in demand in the chemical and surveillance industry. And next thing you know, I came up with a goofy PS to play around.

Now, you would think that this is a typical engineering approach. But as a fresher, the idea of making something cool was way more exciting than solving a real problem. (And for those of you wondering, yes, my robot made it to the orientation video)

Wall Climbing Robot

Designer | Third Year, IIT Madras

The ‘Product Design Lab’ course showcased the designing approach in students. The main objective was not to develop a new product but to come up with a methodology for a desirable outcome.

‘The degrading air quality of urban cities in India’ seemed like a relatable problem to my team for this project. Planting more trees is a no-brainer, or so it seems. But the feasibility and implementation is one daunting task. Our quest as designers to facilitate individuals to plant more trees thus began. Initially, we surveyed several urban dwellers and queried the incumbent issues of setting up a garden indoors. Apart from the few obvious reasons like space and time constraints, installation, maintenance, and plant-sitting are the common roadblocks.

Our solution was a fully automated DIY Vertical Garden coupled with and an engaging community program for fellow gardeners. We acknowledged the implicit requirement for aesthetic and added neon lights & glass façades to enhance the appearance of the garden.

DIY Vertical Garden

Data Scientist | Fifth Year, IIT Madras

Just when we thought we had mastered the art of product design, the department came up with a special provision to test our learning. During the second half of our penultimate year, we are required to work on an industrial project within India. I joined Godrej and Boyce in Mumbai, along with three other batch mates to lay the foundation for products in the healthcare division of Innovation and Design Centre of the company. Halfway through, we realized that the topic we initially picked was too extensive. So, a few brainstorming sessions later, we decided to narrow it down and settled on fitness wearables. 

Delivering the finished and functional prototype was where the project ended for many. But for me, it was the dawn of something new and exciting.

Final year students get a year to work on a master’s thesis. My desire to explore wearables made me pick “Analysis of Pulse signals” as my research topic. Prior experience at Godrej and Boyce had given me a basic understanding of signal processing and machine learning, but this project required a thorough grasp of the topic. When it comes to fitness wearables, the requirement to make a product desirable goes a step further, and the power of data science comes into the picture.

As I buckled up to dive deep, I realized that both technical and domain knowledge plays an important role when it comes to analyzing biomedical signals like PPG and ECG. The sheer number of factors that can influence the behavior of these signals is beyond fascinating. Although much of the burden of designing complex algorithms and machine learning models are taken care of by advanced python libraries, the real challenge lies in the choices one makes.

To conclude, I’d say that it was data science that chose me!  I found my journey to be like a progressive evolution in my skills, as well as the thought process that is essential for my projects. The chic title below my name in the visiting card may vary from project to project, but it sure doesn’t limit my engagement. I believe limits are limited by our definitions.

If we approach life as an optimization problem then the cost function is time dependent. The global optimal solution is unknown and shifting. So, don’t get too comfortable with the current solution as coefficients of the function are perpetually changing.

EDITOR – Ankur Tiwary, Akansha Verma

%d bloggers like this: