real-time analytics
As health and retail sectors become more data-intensive, traditional analytics systems are struggling to keep up. Real-time insights are no longer optional-they're essential. From tracking patient vitals to predicting retail trends, businesses now face increasing pressure to react in the moment. At the same time, rising expectations for privacy, compliance, and reliability are forcing organizations to rethink their infrastructure from the ground up. The push for IoT integration and edge analytics is more than a trend; it's a shift that redefines how data is captured, processed, and used.
Instrumental in this shift is Selvakumar Kalyanasundaram, a data and cloud architect whose work is quietly powering real-time decisions at some of the world's most demanding organizations. With deep experience in healthcare, retail, insurance, and wealth management, he is leading critical efforts to build future-ready, IoT-integrated data pipelines. His current role at CVSHealth has placed him at the center of one of the most complex data modernization journeys in the U.S. healthcare system.
"I've always believed architecture should enable, not limit, the pace at which organizations move," he says. His belief has had a profound effect on the real world. He oversaw the transition of key systems from Hadoop to Google Cloud, which resulted in a 30% decrease in infrastructure expenses and a 45% improvement in analytics performance. These were more than just technological triumphs; they opened the door to the scale and speed required for the analysis of patient data in real time.
Selvakumar's approach combines technical mastery with a strong grasp of real-world constraints. "You can't have downtime when you're dealing with health records or active retail systems," he explains. That principle shaped his leadership on the Integrated PA project, which allowed real-time cross-benefit claims analysis by integrating datasets across pharmacy, benefits, and customer interaction systems. The system not only improved analytics speed, but it also generated millions in healthcare savings.
His work in retail follows the same playbook: make data accessible, reliable, and fast, without breaking the system. He has built a Data-as-a-Service platform that harmonizes customer data for marketing and compliance. The result was a 35% improvement in personalization accuracy and significant gains in e-commerce sales. "We gave marketing teams the ability to act on insights in real time," he recalls. "That kind of agility wasn't possible before."
He then turned to a more complex challenge-bringing together operational data from pricing, HR, and supply chain systems into a unified framework. The result, the Enterprise Data Fabric, cut reporting latency by 60% and brought measurable improvements in operational precision. In every case, his goal remains the same: eliminate delays, increase trust in the data, and enable smarter decisions; whether at the bedside or the checkout counter.
But even the best systems don't build themselves. He has had to solve thorny issues that many shy away from. "One of the hardest problems was executing migrations without interrupting service," he says. "When we moved terabytes of healthcare data from Hadoop to Google Cloud, data integrity and compliance were front and center." He implemented metadata-driven governance frameworks to ensure consistency across datasets and maintain privacy standards like HIPAA and CCPA.
In high-velocity environments, like retail IoT, he's had to be even more creative. He redesigned pricing analytics workflows to reduce processing time from 24 hours to just 2. "The business could suddenly make pricing decisions in near real time," he says. "And that changed how fast they could respond to the market."
Selvakumar's leadership extends beyond technical projects. Though much of his work remains internal, he's authored whitepapers on cloud migration and IoT-driven analytics, and regularly contributes to executive discussions on data modernization. His insights have helped define long-term strategies in both retail and healthcare.
Selvakumar believes the next frontier lies in real-time, predictive intelligence at the edge. His experience across industries has led him to a clear vision for the future of analytics and architecture. "In healthcare, IoT-enabled real-time patient monitoring will be the norm, integrated across payer and provider ecosystems to predict and prevent health crises before they escalate," he says.
The same transformation, he notes, is already underway in retail. "Edge AI will redefine personalization by analyzing customer behavior directly at the device or store level. It'll cut down decision latency to milliseconds and change how businesses respond to individual preferences."
But speed and intelligence must come with responsibility. "As privacy regulations like HIPAA and GDPR evolve, stronger metadata governance will be essential," he adds. For Selvakumar, the future belongs to hybrid edge-cloud architectures, where device-level processing works in tandem with cloud-based aggregation and predictive insights. "That's the only way to scale while staying compliant, efficient, and agile."
Drawing from years of architecting mission-critical systems, he offers a final insight: "Serverless cloud solutions, infused with machine learning, are the key to unlocking scalable, future-ready data pipelines. But the real differentiator is building them with purpose, designed not just for performance, but for trust."
And that, he believes, is how data architecture will continue to shape industries, not just as a backbone of operations, but as a strategic driver of health, retail, and beyond.
