Model's successes cascading to millions: Hatim Kagalwala discusses precision, responsibility, and practical machine learning
In the dynamic world of fintech, Hatim Kagalwala has made significant strides in shaping credit models for emerging markets. His journey began at American Express, where he contributed to the Comprehensive Capital Analysis and Review (CCAR) process, developing forecasting models for credit card spending volumes and paydown rates over a 13-quarter horizon.
One of the major challenges in these markets was behavioural, as consumers often respond differently to credit products compared to those in more developed economies. To overcome this hurdle, Hatim's team utilised alternative data sources such as mobile top-up behaviour and delivery reliability to estimate creditworthiness.
In a business context, causal models help prioritise what should be done to achieve the best outcomes, rather than just predicting what is likely to happen. This approach was instrumental in ensuring American Express maintained sufficient capital buffers to withstand severe economic downturns, contributing to the company's financial stability.
To tackle the lack of reliable data, Hatim's team used ranking approaches to provide a more flexible and nuanced way to prioritise decisions. They also explored various machine learning techniques that could effectively learn patterns from alternative data while maintaining interpretability and fairness.
In the e-commerce space, data science allows for greater experimentation, with a higher tolerance for failure during early development. This was evident in Hatim's work at Amazon, where there was a clear focus on expanding digital commerce in emerging markets, where traditional credit systems are either limited or unreliable.
As the fintech industry continues to grow, data science applications have direct financial consequences for individuals, necessitating explainable, fair, and compliant models. Institutions that take regulatory stress testing seriously are better equipped to navigate uncertainty and maintain trust with regulators, investors, and customers alike.
Hatim's transferable data science skills have been crucial in moving between industries. His risk-aware mindset has proven valuable in fast-moving environments, while his experimentation-driven approach has been a boon in more regulated settings.
Recent expansions into payment options markets in emerging economies include companies like PayPal, Square, and Stripe, motivated by large unbanked populations, digital economy growth, and increased mobile penetration. However, these companies face challenges such as regulatory compliance, infrastructure limitations, and local competition.
In conclusion, Hatim Kagalwala's work in shaping credit models for emerging markets has been instrumental in navigating the unique challenges these markets present. His contributions have helped businesses like American Express and Amazon maintain financial stability and expand into new markets, while also setting a standard for ethical and compliant data science practices in the fintech industry.
