2021-01-27 15:27:27

Till about a decade ago, if you specialised in mathematics, you would end up being a teacher or a researcher. That has changed dramatically. Specialists in the field in India are now among the most sought after in the financial and technology world. Many of today’s leading digital technologies – artificial intelligence, machine learning, data science, big data, cyber security – need strong foundational knowledge of mathematics.

“Data science is based on three skill sets – a background in math or statistics, exposure in computer science, and business or domain knowledge,” says Ashok Kalidas, head of data science & innovation at research firm Kantar’s analytics practice. “Where a math graduate adds value is in the ability to understand the mathematics behind the models and innovate on top of that.”

Kalidas says AI and deep learning solutions can be implemented using software, but you need mathematics to understand the inner workings of these solutions. “A mathematics graduate will be able to interpret why you are getting certain types of results, whether it makes sense, how to modify some of it and go to the next level,” he says. Only mathematics will tell you under what circumstances a solution would work and in what circumstances it would not.

Swaminathan Padmanabhan, senior director of data science at software firm Freshworks, says that mathematics helps create unique and more effective ML models than those built using offthe-shelf libraries and automation platforms.

Chakra Mantena, MD and head of technology at Morgan Stanley India, says the company hires mathematics students in the fields of financial modelling and quantitative roles, which involve sitting with traders and helping them decide what to buy or sell.

“The products we help clients trade in are complex fixed income and derivative products which involve pricing them and projecting cash flows. And when you are doing risk analysis, how a certain trade may play out, some of the math involved is intense, and an engineer may not have had exposure to it,” he says.

This means topics such as probability, functional analysis, topology, algebraic geometry, number theory, and graph theory assume great importance. “People who have knowledge in algorithms, probability, linear algebra, statistical methods tend to get absorbed in the finance sector and research labs,” S Dharmaraja, head of the department of mathematics at IIT Delhi, says.

The growing need for math professionals is reflecting in developments at India’s leading educational institutions. The student intake at IIT-Madras’s mathematics department has more than doubled to 30 now, from 12 in 2018. Chennai Mathematical Institute (CMI), which counts Ford, Crisil, Adobe and Credit Suisse among its recruiters, has seen pay packages double to Rs 15.5 lakh this year, from Rs 7.7 lakh six years ago.

“This year, the number of applications increased by almost 60% – to 1,200, from around 700 in the previous year – for the course which is focused on providing jobready skills in data science,” Madhavan Mukund, deputy director and dean of CMI, says.

Salman Abdul Moiz, chairman of the placement guidance and advisory bureau at the University of Hyderabad, points out that until 2016-17, a majority of the students from the school of mathematics and statistics would pursue careers in academics, but in the last two to three years, around 60% of the students are getting recruited by tech firms.

Anupam Kaura, president of HR at ratings & research firm Crisil, says a degree in math is important for his firm because the quantitative, statistical and analytical skills are applied to design, review and refine complex time series models to provide insights into business problems of customers, including regulatory compliances.

“Given that you need to look at a lot of data structures, run statistical tests, regression tests, these are roles for which an ISI (Indian Statistical Institute) graduate is more skilled than an engineer,” he says. Crisil usually hires annually from ISI Kolkata, CMI, Centre for Modelling and Simulation – Pune, Indira Gandhi Institute of Development Research, besides the top IITs.

Many educational technology companies are hiring those with Master’s in math not just for teaching math but subjects like data science concepts, computer science algorithms, applied mathematics. “If you look at data scientists in research and in MNCs in technology and finance, they are all PhDs in mathematics,” says Kaushik Banerjee, VP and business head of staffing firm Teamlease.

Cloud software firm Zoho has hired pure math freshers to its research team for AI & ML-led projects. Shailesh Kumar Davey, co-founder and director of engineering, says such talent helps the engineering team by translating complex research concepts to simpler terms, thereby helping the latter develop good code. “Digital marketing roles also require candidates with knowledge of stats & math,” he says.

India needs a lot more of such skills. Neeraj Sharma, VP of HR at logistics software platform FourKites, says despite hiring from the top statistics schools, they face a skills gap for analytics roles. Data science jobs today, he says, require a knowledge of both statistics and computing techniques.