.png){kind=logo}
With over 12 years of experience in financial risk modeling, Piyush Sinha, Director of Data Science at Marsh McLennan, has established himself as a leader in AI-driven risk analytics. His expertise spans machine learning, generative AI, synthetic data generation, and advanced modeling techniques such as neural networks, boosting, bagging, and regression models. Throughout his career, Piyush has played a key role in integrating AI into financial risk management, helping organizations enhance decision-making and navigate complex risk landscapes.
As a leader, Piyush believes that the foundation of an effective team lies in vision, communication, and empathy. He sees leadership not as a mere oversight role but as a responsibility to help teams understand the larger impact of their work.
“A leader must ensure that every team member knows how their work connects to the broader business goals. This clarity not only improves project outcomes but also creates a sense of purpose within the team.”
For Piyush, open communication is key. He encourages regular one-on-one discussions, team meetings, and project-specific check-ins to align expectations and resolve differences. Transparent communication fosters trust, allowing team members to feel valued and empowered in their roles.
Empathy also plays a significant role in his leadership style. He believes that understanding different perspectives and motivations within the team leads to a more collaborative and innovative work environment. By acknowledging the human aspect of decision-making, he creates a culture where individuals feel supported, heard, and motivated to push boundaries.
Piyush has worked extensively in banking and insurance risk analytics, fields that have undergone a massive transformation in recent years. Traditionally, financial institutions relied on statistical models that were heavily regulated and included subjective decision-making. However, as data availability and computing power have grown, businesses now leverage alternative data sources to assess risk more accurately.
“Risk modeling has evolved from simple statistics to advanced AI models. Today, banks assess credit risk using unconventional indicators like satellite images of properties, telecom usage patterns, and social media sentiment. The insurance sector is also incorporating stock market trends and company financials into their risk models.”
At Marsh McLennan, Piyush has been instrumental in developing an AI-powered risk analytics suite, which has digitized traditional risk assessment frameworks. His contributions have helped clients quantify and benchmark their risk against industry peers, providing them with deeper insights into their exposure and mitigation strategies.
Piyush attributes much of his success to his ability to adapt to change. Over the past decade, data science and AI have evolved rapidly, demanding constant upskilling and innovation.
“When I started, we used basic tools and had to write lengthy codes for every task. Processing large datasets was cumbersome, and insights were limited. Now, open-source platforms like Python and R, coupled with cloud computing, have transformed how we handle data. The industry has moved from predictive analytics to machine learning, and now to generative AI and LLMs.”
He has consistently embraced these changes, whether by learning new algorithms, refining his ability to communicate data-driven stories, or improving his ability to influence business decisions. His commitment to continuous learning has allowed him to stay ahead of industry shifts and bring innovative solutions to the organizations he has worked with.
Beyond technical knowledge, Piyush also recognizes the importance of storytelling in data science. Early in his career, he focused primarily on delivering models and reports. However, as expectations evolved, he realized the power of narrative in influencing senior leadership and shaping business strategies.
“Effective communication is just as important as technical expertise. Turning complex analyses into compelling stories ensures that insights drive meaningful action.”
Influence of Ancient Wisdom on Decision-Making
A significant influence on Piyush’s professional and personal philosophy is India’s ancient concept of karma, which emphasizes focusing on actions rather than fixating on outcomes.
“This philosophy has shaped the way I approach my work. By prioritizing effort, skill-building, and execution, results naturally follow. It removes unnecessary anxiety over things beyond our control and helps us stay focused on what truly matters.”
This mindset has guided him through uncertainties, challenges, and high-pressure decision-making, enabling him to lead with clarity and resilience.
While AI and risk analytics have advanced significantly, the industry still faces critical challenges that require careful navigation.
Risk events such as loan defaults and insurance claims occur infrequently, making high-quality risk data limited. This poses a significant challenge in building predictive models.
Understanding business context is just as important as technical expertise. Piyush believes that the best data scientists are those who take the time to deeply understand how business processes work.
“The most valuable insights come from truly understanding the data behind business operations. Eliminating variables without knowing their significance can lead to flawed models.”
While AI has made risk analytics more efficient, businesses still hesitate to fully adopt AI-driven insights. Piyush sees his role as an advocate for AI adoption, ensuring models are not just technically sound but also trusted by stakeholders.
One of the toughest challenges in AI is proving its impact. While AI-driven insights add tremendous value, measuring their direct effect on revenue, efficiency, or risk mitigation is not always straightforward.
“AI success isn’t just about building great models; it’s about demonstrating real-world impact. Business leaders want to see how AI translates into cost savings, efficiency, or competitive advantage.”
As AI adoption grows, concerns about data quality, fairness, and bias become more critical. Piyush believes AI developers must prioritize ethical AI practices to ensure transparency and accountability.
Piyush envisions a future where data science is democratized, much like photography or video editing, which once required specialists but is now accessible to everyone.
“The next wave of AI will empower business users to perform their own analyses, while data scientists will transition into the role of enablers—developing platforms that make AI insights more accessible and actionable.”
However, he warns that trust in AI will be a growing concern. As businesses increasingly rely on machine learning for decision-making, the demand for explainable and ethical AI models will rise.
At the same time, he remains optimistic about AI’s potential to revolutionize risk analytics, financial services, and beyond. His focus remains on ensuring AI is not just powerful but also practical, ethical, and aligned with business needs.
