The world’s leading publication for data science, AI, and ML professionals.

Are you a Data Scientist aspirant? Here is my story of becoming one

I desperately wanted to become a Data Scientist and finally became one in December 2016 after 18 months of self-learning. In this post I…

I desperately wanted to become a Data Scientist and finally became one in December 2016 after 18 months of self-learning. In this post I talk about what I was doing before becoming a Data Scientist, how did I become one and what is it like being one.

Photo by Alexander Sinn on Unsplash
Photo by Alexander Sinn on Unsplash

Before Data Science:

I studied Computer Science and Engineering from a tier 3 college in India, graduated in 2013. I had no clue what I wanted to do in my life and in my career. I got 2 offers during my final year of college. One from HCL Technologies and the other from BNY Mellon. I took the later offer as they asked me to join immediately after my college ended.

So my career in the corporate world as a Software Engineer started in September 2013 where I was trained on Mainframes. Mainframe is a really old technology, invented in the 1950s. They sound cool only in movies where some dude with a headset yells, "they hacked into our mainframes!"(they have never been hacked in reality). I’m passionate about programming so I loved the training as I had lots of hands on. After I was put on a project, the initial few months were awesome as I got to learn the application. Once I got familiar with the app, things became mundane. So I switched to HSBC Technology, India as they promised some pure development work.

Pursuing Data Science:

After joining HSBC I got to know that there is no pure development work on Mainframes anywhere in the world. This was the start.

Around mid 2015, everybody around me was talking about Hadoop. So I thought why don’t I move into Hadoop as this was the buzzword and of course, one of the highly paying job. So I bought a new laptop, followed tutorials on Hadoop and got certified. I did Hortonworks Data Platform Certified Developer(HDPCD).

Again, wasn’t satisfied with Hadoop. I kept on reading about Data Science and did some MOOCs. One thing led to another and I finally found my new love, Machine Learning. Then I concentrated more on ML, enrolled for the ML specialization by University of Washington in Coursera and it was a fabulous course(a little bummed out when they cancelled the last module and the capstone. I blame Apple! ).

After a year and a half of my MOOC journey I felt like I now know some stuff and I can really solve some problems. I then spammed almost all the heads and senior members in my company about my aspirations. I mentioned in the mail that "I’m an Aspiring Data Scientist and I want to practice Machine Learning". Also, I mentioned all the certifications and courses that I have done in the mail. Few of them replied back and few of them directed me to another more relevant person. Never thought I’d get this kind of response from all those big people. Eventually, one door opened for me. The Head of big data, India who also leads the India’s Innovation Lab asked me to meet him the very next day I sent him the email. This was the turning point of my career.

We met in the Innovation lab and he asked me about what I can do. I blabbered a bit about Kaggle(I’ve tried a few problems). I was pretty sure that I messed up the only chance I have ever gotten. But to my surprise, they wanted me to work with them but not officially as at that time there wasn’t any projects on Data Science in India. That was the time when TensorFlow was open sourced and everybody in our bank wanted to see how we could use it. So the head asked me to take a lead on this and do something. It took me almost a week to get my head around TensorFlow and to my luck I got LendingClub’s dataset from Kaggle. Using this I developed a Credit Risk model and presented it to the iLab folks. They all loved it. We then presented it to our senior members and heads of various departments. Everybody loved it. Guess this gave people confidence in me that I’d be able to do the job.

To my luck, a job in Data Science for our Global Markets division opened up in India by December 2016 and the head suggested my name. Was interviewed for the same and was put on the project. I was the first ever Data Scientist in India which at that time had 10K+ employees.

As a Data Scientist:

My first 6 months of being a Data Scientist were the most challenging period in my career. It was a small team, there were 5 of us, and thankfully I had one of the best leads that I could hope for. He taught the real life Data Science as opposed to what I learnt from Kaggle and from doing MOOCs. I struggled a bit in the first few months as the analytics jargon is quite different from software engineering world and also, being very close to business I had to put in extra effort to gain domain knowledge on what I was working. I worked crazy hours, for almost 16 hours a day and over the weekends too! I was learning, I was happy and I was doing something new every single day! Alongside my job, I started a Machine Learning Fight Club (The first rule of the club is to talk about the club!) and started sharing whatever I learnt by organizing sessions.

After 2.5 years into Data Science, I found my current job listed in our Internal Job Listings portal. The job’s location was in Toronto, it was for a new Data Lab that HSBC was setting up. The job looked so appealing for someone who likes research and I wanted to explore other parts of the world. So I applied for the role, went through almost 5 rounds of Interview and finally started working in the Toronto Data Lab from May 19th, 2019. I’m now part of a 30 member global(another lab in London) team with Graduate Data Scientists, Data Engineers and Senior Data Scientists. Being part of a bigger team and also a senior DS I get to practice my soft skills by leading a use case, managing stakeholders, etc. It’s one of the best teams I have ever been in!

Summary:

I wouldn’t say I was born to be a Data Scientist, it all started with being dissatisfied with what I was doing. Getting into Data Science without any credentials was tough but I worked hard for it, kept learning, knocked on any door that I saw and it paid off. The MOOCs assisted me with the right techniques and tools but applying to real problems is something only practice will teach. I’m very happy being a Data Scientist, every day is a new day, there is still a lot to learn which keeps me from getting bored!


Related Articles