Passion for machine learning: Q&A with Nirankar Singh
Nirankar Singh is a quantitative analyst. Before joining Northmill he worked at the Riksbanken; Sweden's central bank, where he was involved in a research program. He has an M.Sc. Applied Mathematics from KTH Royal Institute of Technology.
The rise of the machines in banking
Machine learning is certainly becoming increasingly trendy among the lending and banking industry for building credit risk models and predicting low-quality loans. Providing computers with the ability to learn without being programmed is a driving force of the fintech-revolution. We sat down with Nirankar Singh.
Nirankar, please tell us a little about yourself
My name is Nirankar, I work as a quantitative analyst, a quant, at one of Sweden's fastest growing fintech-companies. Currently, I’m in charge of developing our credit scoring models using machine learning capabilities.
Quants are in great demand. What made you choose Northmill?
I wanted to work for a successful fintech-company with a strong leadership and with a vision to transform finance. Northmill was the perfect match.
What exactly are you working on right now?
My main priority projects are credit scoring models and Enterprise Risk Management (ERM). We have come a long way in testing and implementing different scoring algorithms using machine learning.
Northmill has 10 years experience in credit scoring but this sounds like a big breakthrough?
Yes, and the best part is that we have built the algorithm testing process from scratch. The Time to Market (TTM) for a new model is optimized, meaning we can implement a new model within a couple of hours!
The analyst team
How can financial companies leverage from big data?
I would say it depends on how well you know your customers. Credit bureaus have a vast amount of data but their scoring models are made to fit many markets and can be very general.
To optimize the credit risk, one must create their own credit models that reflect its customer base. We do not have only one but several models for different types of customers. We believe that this differentiation among groups is the best way to minimize risk.
”Credit scoring will to a much greater extent be correlated with handling big data”
How do you see the future of consumer credit scoring?
Credit scoring will to a much greater extent be correlated with handling a big amount of data and, more importantly, how to utilize this data. The ability to squeeze every ounce of information from different sources and using new and improved algorithms is a huge challenge. A challenge which requires knowledge of what kind of data is collected.
I think Northmill have come far in making data available across the company, meaning that unstructured data have been transformed to more easily interpretable data. You can only imagine how this is benefiting the whole organization.
Is there any substance in credit check by social media?
In recent time, social media scoring has begun to gain momentum since the amount of data that is generated on social media is an enormous source of information.
We have been following this kind of alternative scoring approaches for a while and our data infrastructure is built in such a way that when the time is right, we will harvest the benefits from any similar kind of innovative approaches for credit scoring.
Where do you see Northmill within a year?
I have no doubt that we will build the best digital-only financial hub available for private consumers. To achieve this goal, we need to see a growth in the number of quantitative analysts in our department during the upcoming year. This will give us a more competitive advantage by tirelessly going above and beyond customer expectations.
Read more: Top 3 from Fintech Stockholm 2016