Banking Industry Data Science Expert Demonstrates How Machine Learning Drove 600% Growth in Deposit Conversion
In January 2025, Euroclear, one of the two European international securities depositories with over half a century of history, announced a strategic partnership with Microsoft, aiming to introduce innovative technologies, such as cloud computing, data-centric approaches, and AI-based analytics into its capital markets ecosystem. During the seven years of the partnership, the company plans to pursue innovation in several aspects, from enhancing the customer experience to establishing an ecosystem for financial data-sharing and strengthening security and reliability. This is just one of many examples of innovative technologies finding their place in the financial industry. Pavel Baltabaev is an analytics and AI expert in the banking industry working at a premier consulting firm has been one of the first experts to introduce AI automation in lending underwriting earlier this decade and, more recently, pioneered GenAI tools to drive productivity growth in leading financial institutions across the globe. Being among the few early innovators in this space, Pavel believes that the influence of these novel technologies will only expand, making it essential for companies to embrace these advancements to maintain a competitive edge.
From the beginning of the recent AI boom, experts remained wary of its application in finance, as well as other sensitive industries, where the cost of a mistake can be high. However, with the growing number of practical examples, the benefits and limitations of the technology become better researched, and more companies are willing to adopt it. A recent report shows that 72% of finance leaders are utilizing AI in various departments, including risk management and fraud detection.
“On the surface level, AI-based solutions are often perceived as not reliable enough or tied to additional risks,” – explains Pavel Baltabaev. – “However, in practice, it helps to enhance security and create a basis for more productive decision-making”.
For example, Pavel Baltabaev has revolutionized the traditional credit underwriting process by integrating AI solutions that automate risk assessment and decision-making. His innovative approach, first implemented between 2020 and 2022 in Russia for personal loans and credit cards, transformed a process that once took hours or even days into one that now completes in mere seconds for many applications. By connecting diverse internal and third-party datasets, Mr. Baltabaev’s system leverages AI models trained to mirror human judgment, drastically reducing the need for manual intervention. His work significantly lowered the manual underwriting rate from 80-90% to 20-40%, allowing banks to optimize operations and improve customer experience. In 2023, he extended his breakthrough methodology to the US market for home equity loans, cutting processing times from up to a month down to just one week, further cementing his impact on the financial industry’s digital transformation.
AI applications in banking extend well beyond credit risk management, with Pavel Baltabaev highlighting customer value management and workforce productivity as two of the most promising areas. In 2024, he implemented a customer-value optimized approach by developing a deposit cross-selling machine learning model that takes into account both conversion and balance size for a major Central Asian bank, resulting in a remarkable 600% increase in conversion rates and generating an additional $35 million in client deposits. That same year, Baltabaev pioneered one of the first generative AI knowledge assistants in the banking industry for an international investment bank. This innovative tool empowers bankers to quickly access insights from previous deals, summarize extensive documents, and prepare pitch decks 20% faster.
However, the quality of the results that can be achieved through applying AI technologies heavily depends on the quality of the data that are used as a basis for the model. It also requires an understanding of the specifics and limitations of the AI tools used in particular projects.
“The AI cannot be mindlessly applied to any issue at hand,” comments Pavel Baltabaev. “For it to be efficient, tailored solutions are required, which will account for the specifics of the particular organization or segment.”
This is why professionals who have a strong background in both computer science and finance are in increasing demand. Pavel Baltabaev mentions that studying mathematics, statistics, and programming at the Higher School of Economics (HSE), one of the leading research universities in Russia and Eastern Europe, ranked as No.1 by Forbes among the most reputable universities among employers in Russia and possessing the highest ranking of the quality of AI education in Russia according to an independent evaluation. Studying at HSE provided him with a strong educational foundation. It helped him to develop structured thinking and a deep understanding of modern ML and statistics, later applying this knowledge to business and consulting projects. However, it remains important to learn continuously, as technology is constantly evolving.
“The transition from more straightforward and simplified models to generative AI and large language models took several years, and there are no signs of the developments slowing down,” concludes Pavel Baltabaev. “New technologies emerge every year, and one needs to explore them and find the ways to apply them to not be surpassed by competitors.”
This is why Pavel finds it important to create and support an environment that encourages professional improvement and continuous learning. For example, in 2025 he participated in the panel of judges at BrowserBuddy Hackathon, organized by Hackathon Raptors a prestigious association that includes outstanding software developers and engineers who made a valuable contribution to the community.
He adds that the finance industry will continue to evolve, and new technologies are already becoming an integral part of it. There is hope that the changes brought by them will remain mostly positive both for financial organizations and their clients, ranging from improved security and better decision-making to improved customer experience and personalized offers.