As the midpoint of 2025 approaches, Amazon Web Services Inc. is sending a clear signal: Software development is undergoing a fundamental repositioning with agentic AI at the helm. The typical, grunt work-reliant modus operandi is giving way to a spec-driven development approach that democratizes entry and rewards ingenuity and versatility.

AWS’ Deepak Singh explores Kiro’s advantages for spec-driven development.

AWS’ vision? A future powered by new frameworks and powerful tooling, redefining what it means to build software. The unveiling of Kiro, AgentCore and Model Context Protocol serves as the first step in concretizing that end goal, according to Deepak Singh (pictured, left), vice president of developer agents and experiences at AWS.

“This whole idea of spec-driven development and specifications started shining and we started testing this out with the customers,” Singh said. “Eventually, what we ended up with, and what people were doing, was they were handcrafting sort of these spec-type workflows within whatever they were doing. With Kiro, that’s how you work — it’s built into how Kiro works.”

Singh spoke with theCUBE’s John Furrier (right) at theCUBE + NYSE Wired: AI + Cloud Leaders Media Week event, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed spec-driven development and the new toolkit from AWS aimed at realizing that paradigm.

Kiro simplifies spec-driven development with agentic AI

At its core, Kiro has been designed to allow users to alternate between vibe coding and structured design. Purpose-built for devs creating with agents, users can collaborate with agents to define application requirements, structure functionality and scale from prototype to production — all within a flexible AI-native workflow, according to Singh.

“You can have the agent write specifications for you, and you collaborate with the agent to figure out the requirements and application structure,” he said.  “As the application goes from your little fund prototype to a full application, it grows with you. The customer response in the last 24 to 48 hours has been phenomenal, and we’ll see where this goes.”

Unlike traditional integrated development environments, Kiro integrates agents directly into the development process. In essence, it’s a system that grows with the user.

Complementing Kiro is AgentCore, a new set of primitives within Amazon Bedrock that allows developers to build and run agents. These include runtime environments, browser runtimes, agent telemetry, identity, code execution sandboxes and more, Singh explained.

“It’s all the things that over the last two years, as we’ve been building our agents, that we have learned that are needed to build a successful agentic system,” he said. “That’s what AgentCore is. In many ways, you can think about it as everything, the toolbelt and the services that any individual or company needs to build a very successful agentic software platform.”

The system abstracts complexity away from developers, enabling them to offload logic from LLMs to tools, boosting performance, reducing token costs and increasing reliability through deterministic tool calling. AgentCore is also framework and model-agnostic, supporting popular tools such as LangChain, CrewAI and Strands, ensuring developers can remain flexible as the ecosystem evolves, Singh added.

“Now you can just call the API, which is available through an MCP server that you’re making available as a tool to the agent,” he said. “That’s radically changing how some of these agentic architectures work. It’s a great time to be part of this growth.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of theCUBE + NYSE Wired: AI + Cloud Leaders Media Week event:

Photo: SiliconANGLE