Sunil Kumar Paidi

From APIs to AI Agents: How Model Context Protocol is Reshaping Enterprise Software Architecture

Abstract:

We have been building APIs for over two decades. REST worked. GraphQL solved real problems. gRPC made things faster. But none of them were built with AI in mind — and that gap is starting to show.

As AI agents move from research labs into production systems, enterprises are discovering that their existing integration infrastructure simply was not designed for this. Agents need context. They need to discover tools dynamically. They need to act across multiple systems in a single reasoning loop. Traditional APIs handle none of this gracefully.

Model Context Protocol (MCP) is an open standard that gives AI models a structured, secure way to connect with the tools and data they need to get work done. This session covers how MCP is architecturally different from what we have used before, shares real examples of how teams are deploying it today, and takes an honest look at where the challenges still are.

Brief profile:

Sunil Kumar Paidi is a Lead Software Engineer specializing in backend systems, distributed architectures, and AI-driven intelligent platforms and extensive experience building scalable applications using Java, Spring Boot microservices, and cloud-native technologies. Mainly focuses on integrating artificial intelligence with large-scale backend systems to enable intelligent data retrieval, real-time analytics, and recommendation engines. I have experience in designing and integrated microservices across multiple enterprise systems while improving system performance, scalability, and reliability