The software industry is racing toward agent-driven systems powered by artificial intelligence. To optimize and scale them, companies are standardizing on robust infrastructure and on the Model Context Protocol with the hope that MCP will simplify integration across tools and data sources.

Agents are rapidly becoming the next generation of integration platforms, connecting more tools and data sources across entire enterprise environments. They’re replacing older, heavyweight systems with a more flexible, intelligent foundation that acts as the new nervous system of modern business operations, according to Marco Palladino (pictured), co-founder and chief technology officer of Kong Inc.

“I’m very bullish on agents. I’m very bullish on the ability to generate agents in a quicker way by abstracting away the infrastructure requirements, and I’m very bullish about exposing some of these agents to the customers that are now starting to use our products through the AI prompt,” Palladino said. “You could think of the AI prompt as being the smartest browser in the world.”

Palladino spoke with theCUBE’s Paul Nashawaty at the Kong API Summit: The API Summit for the Agentic Era, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed how AI is reshaping enterprise infrastructure and the rise of agentic systems powered by new protocols like MCP. (* Disclosure below.)

Production-ready MCP deployments

Recently, Kong unveiled a new enterprise-grade MCP Gateway to ensure enterprises are able to roll out production-ready MCP deployments. MCP has rapidly emerged as a foundational technology for agentic systems and AI applications, according to Palladino. While Representational State Transfer (or REST) has long been the dominant architectural style for traditional application programming interfaces, the agentic era requires a native protocol for tool use and context exchange.

“MCP represents to AI what REST represented to traditional APIs,” he said. “It is a new protocol that enables agents to be able to communicate with third-party data sources, third-party services – just like an API.”

It operates in an agentic-native way, making real-time communication the standard. This approach also enables the discovery from the agents by default, according to Palladino.

“It is fundamentally a modern protocol for building agentic AI,” he said. “Of course, MCP, like every other API that we have, also needs to be governed, needs to be secure, needs to be observable.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of the Kong API Summit event:

(* Disclosure: TheCUBE is a paid media partner for the Kong API Summit. Neither Kong, the sponsor of theCUBE’s event coverage, nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE