Research & Development requires a multi-disciplinary lens, with a background in software – programming, data science – ML / Math and engineering, coupled with domain knowledge.
Akshat Rastogi, our next pathbreaker, Associate Data Scientist – II at Philips Innovation in Bangalore, works on building digital twins as part of an advanced development initiative.
Akshat talks to Shyam Krishnamurthy from The Interview Portal about applying deep learning in conjunction with numerical modelling to solve industrial problems as part of Industry 4.0, and the potential impact of these technologies in bringing about meaningful innovations.
For students, take the word “passion” with a pinch of salt. Life is more like a multi-objective optimization problem, that too with varying constraints with time. Keep exploring and exploiting opportunities until you find mental peace, challenge yourself and are able to spend time with family / friends !
Akshat, Your background?
I always face difficulty in answering the question – “Where are you from?”
In a nutshell, my roots are in Punjab and UP, because my grandparents / parents hail from there, though they were brought up in different cities. I was born in Hyderabad and did my schooling from 5 cities across India – Ludhiana, Bangalore, Hyderabad, Vapi / Daman and Baroda. So, I am basically a North Indian, although my upbringing and formative years were in South and West India.
I was interested in asking questions from a very early age and started to drift towards Science and Mathematics. Since we did not have access to the internet at the scale at which we have it since the last 10 yrs, our means of acquiring interests or even realizing what we liked was limited. In technical terms, the space for finding or exploring what one liked and is good at was restricted to our neighbourhood or physical life experiences or based on someone’s advice.
So, playing around with Hot Wheels, GI-JOE, watching action and sci-fi movies, I got fascinated with cars, aeroplanes, military and technology.
In secondary school, I got exposed to coding and enjoyed the process of building logic to solve problems.
What did you do for graduation/post graduation?
My formal education is BTech in Mechanical Engineering from NIT Surat and MSc in Computational Mechanics (a combination of mathematics, computer science and physics) from University of Stuttgart in Germany.
What made you choose such an offbeat, unconventional and uncommon career?
After brief stints during internship / full-time jobs in Sales, Teaching, Creative roles and Entrepreneurship, I realised that one can do pretty much anything he/she wants to do, but the question that arises is, do you really want to and when do you want to pursue it?
Also, the dilemma is if you like to do many things, how do you choose or realise what you really like doing that not only pays the bills but also aligns with your core values. So, as a result, in our 20s, we welcome the curse of dimensionality or choices and honestly, it’s tough. I still feel the majority of us don’t really know what we are doing with our lives. Let’s normalise this.
I’m currently working as a data scientist as it aligns with my current state of mind / likings i.e, a role that is explorative, challenging, technology oriented, has the potential to drive change by utilizing my domain knowledge acquired through previous experiences and which also pays the bills.
My interests may change in the future as the external world changes, circumstances change, priorities change, or new opportunities may arise for me to grow professionally.
Hence, my career path has not been along a linear line as it is probably for most of the folks who graduated post 2010.
How did you plan the steps to get into the career you wanted? Or how did you make a transition to a new career? Tell us about your career path
I started my career with Siemens as an Executive (Design Engineer) in the Energy Sector, where I got exposure to basic and detail design engineering of power plants and got the opportunity to work on international projects.
Since I wasn’t completely sure of going for an MBA or MSc, I also tried my hand at technical sales / consulting at Apollo Tyres where I was part of a special group to create a pull action in the market. We were tasked to acquire the fleet data, analyse them to derive valuable insights and help us take appropriate action that would lead to increase in overall sales.
Having gained experience in both a technical and a non-technical role within good companies, I was more inclined to strengthen my knowledge in the technical vertical and decided to pursue MSc in Computational Mechanics to understand the “Why” of a design.
While pursuing my MSc, I wanted to understand more aspects of software engineering / CAE (Computer aided Engineering) software and got an internship at Dassault Systems, where I got an opportunity to understand and implement functionality related to topology optimisation and evaluate proof of concepts.
It was during this time, AI was just getting started to be applied in different areas and I saw its potential in applications with numerical simulation. Fortunately, I got a chance to do my thesis in this direction at TRUMPF. I built a proof of concept demonstrating how deep learning can be used in conjunction with numerical simulation to solve an industrial problem as part of Industry 4.0.
In order to follow my curiosity in combining simulation with AI and work within industrial research, I applied to Fraunhofer SCAI. There, I worked on extending an in-house ML algorithm and applying it to an industrial use-case. The work was extremely exciting and challenging.
At Yanxiki, I continued to work on another exciting problem to build a surrogate / ML model to perform multi-objective optimisation problem of an aerospace component.
How did you get your first break?
Despite having graduated from an institute of national importance, the joining date of my first job was deferred by almost 6 months to a year. This was unexpected and that’s when I realized life’s about to get tough and destiny cannot be ignored. My first break for a paid internship was at CIIE, IIM Ahmedabad, and was through effort and not networking. The full time job at Siemens was also through efforts. It depends on what you believe in – efforts, networking, efforts + networking. Something would click and that’s what we may call luck or success.
Where do you work now?
I’m currently working with Philips Innovation in Bangalore. We are trying to build digital twins as part of an advanced development initiative. The use-cases need to be explored and identified based on the industry one is operating in and the available data.
We are trying to detect failures of components accurately and also exploring if we can predict them in advance.
The skills required depend on the projects one is working on. It is crucial to identify the skills that need to be learnt. The skills required are at the intersection of software engineering – programming, data science – ML / math, engineering fundamentals – domain knowledge and business understanding.
I believe my career trajectory has helped me to acquire these skills gradually. Good things take time and effort.
How does your work benefit society?
Being part of Philips is a matter of pride, because they genuinely have a purpose, i.e. to make life better using meaningful innovation.
Our work currently helps to predict any potential problems arising inside an MRI machine at a hospital across the world and take appropriate action proactively. This in turn helps the radiologist, technician and eventually the patient’s experience.
Tell us an example of a specific memorable work you did that is very close to you!
My MSc thesis at TRUMPF Gmbh in Stuttgart, Germany is a memorable piece of work apart from other work that I chose to be a part of.
Your advice to students based on your experience?
I don’t believe in the word ‘passion’ and all sorts of jargons fed via social media, because the more you read about them the more perspectives you get and ultimately it confuses you to the core depending on your age, experience, thinking ability, strengths / weakness, financial flexing ability, etc.
Life is more like a multi-objective optimization problem, that too with varying constraints with time. So, depending on what’s your objective for the next x years (finding this or fixing this is tough though), keep exploring and exploiting opportunities until you find mental peace, challenge yourself, are able to spend time with family / friends, do something out of work hours that doesn’t pay the bills, get time to create Instagram worthy stuff, build a healthy life, etc. This purely depends on what kind of life you want to live and it may change with time or circumstances.
That was inspirational Akshat!