Linear Squared are streets ahead when it comes to Big Data and AI. Their clients are as diverse as Dialog and Brandix, and they are only two years old. Roar Media caught up with co-Founder and Chief Data Scientist Sankha Muthu Poruthotage, over a live email conversation, to find out how a PHD in Statistics led to him founding the largest specialised data science company in Sri Lanka, with a multi-million dollar valuation. Just don’t ask Sankha what Linear does…

So tell us, what does Linear Squared do…?

Linear Squared provides end-to-end data driven solutions to complex business problems leveraging machine learning and artificial intelligence.

Why don’t you tell us something about your background?

I went to school at Nalanda, did my undergraduate degree at the University of Colombo and started my career as a merchandiser at MAS. However, I quickly shifted to analysing data for a living. I was at Nielsen as a manager of quantitative market research and then at Amba research (now, Moody’s Analytics), as an AVP of quantitative financial research.

Sankha received his PHD in Statistics from the University of Connecticut in 2013. Image courtesy University of Connecticut

And what did you do next?

Around this time I realised that I needed to learn more if I wanted to do something innovative in the quantitative side. So my wife and I went to grad school. I was at the University of Connecticut, USA from 2007 to 2013—a bit too long when I look back at it! But I loved it: teaching, research, travelling, getting to know people from all over the world…it was fun. I earned my Ph.D. in statistics in 2013 and worked for a large insurance company in Boston in advanced analytics.

That sounds great. What brought you home to Sri Lanka?

Good question. By this time I was spurred by the entrepreneurial spirit that exists in the DNA of the USA. But my wife and I wanted to come back to Sri Lanka and start something here. Linear Squared was formerly incorporated in March 2015 with my good friend, Rajith Munasinghe, as the co-founder.

When did you figure out that big data was going to be an industry trend?

Well, by the time I graduated in 2013, it was clear that big data was going to be the next big thing. But because of my PhD in Stats, I had the advantage of getting into data science well before the big data “hype”. Luckily, I had a very good idea of what it takes to solve real business problems using data, whilst most companies, even after a half a decade of intense hype and speculation, still seem to think they can have good data driven solutions just by investing in a fancy data platform or an analytics platform. Some even hire fresh undergrads from quantitative backgrounds and expect them to solve some of their most difficult problems!

The Linear Squared Squad. Image courtesy Daily News

So what’s the big (data) secret at Linear?

For an effective data driven solution, three important components need to be seamlessly aligned: 1) technology 2) sophisticated analytics and 3) domain knowledge. Most other companies don’t have these capabilities in-house, or they are misaligned. We at Linear Squared bring all these capabilities on to the table. We are essentially a collection of mathematicians, statisticians, data scientists, engineers, and domain specialists working in perfect harmony to come up with an amazing set of data driven solutions.

You make it sound easy, but it couldn’t have been. What about challenges?

Trust me—nothing was easy! Even getting a water dispenser was a challenge at the start! But I think these teething problems are common and just part of the journey. The biggest challenge though is the “data science skill”—which is our core strength. This cannot be bought off the shelf. In that sense, what we do at Linear Squared is very different from software engineering. We mould data scientists out of statisticians, mathematicians, and engineers. It takes some time.

And how did you attract the big investors?

We were lucky, I’ve been in touch with Mr. Rajan Anandan from Google for some time. Eventually that discussion materialised into a round of investment by Blue Ocean Ventures, a Singapore-based VC headed by Rajan and Mr. Prajeeth Balasubramanium. My belief is that the best way to attract investors is by doing something attractive!

Linear’s Big Data process moulds scientists out of statisticians, mathematicians, and engineers. Image courtesy University of Warwick

So, what is next for Linear Squared?

We mainly focus on four sectors: apparel, telecom, finance, and FMCG. We are trying to solve some of the big problems that are prevalent in these sectors. If we can solve them well, in a scalable way, there is no reason why we should be restricted to Sri Lanka.

We also want to keep building our team, not just in terms of numbers but also with knowledge and skill. That will be the key if you want to compete in the global market. We need to form better collaborative partnerships with our higher education institutions. It is not a coincidence that Silicon Valley is surrounded by some of the world’s best technology universities such as Stanford, Caltech, UCLA and UC Berkley. It is a prime example of industry and academia feeding off each other, enriching both.

Finally, what advice would you have for startups out there?

Well, I would like to say something about acceleration. Acceleration is a must for any start up, no question about it. At the same time, entrepreneurs need to make sure that the business and the rest of the team is ready for it. Think of it as a car. Acceleration is what creates all the noise and attention, but make sure the other parts of the car are ready to withstand the intensity.

That’s an unexpected metaphor!Thanks for your time Sankha, and best of luck to everyone at Linear Squared.

Thank you Team Roar—we love the website.

Cover image courtesy: