Oracle Corp. today announced a new artificial intelligence-native experience for its Fusion Applications within Oracle AI Agent Studio, allowing developers, customers and partners to build agents together.

This expansion fills in the developer side of Oracle’s Fusion app agentic enterprise strategy. The company’s AI Agent Studio initially allowed business users and partners to assemble agents with natural language and low-code tools, while Fusion Agentic Applications turned teams of those agents into autonomous, outcome-driven software.

“[We started with] people who can use natural language or a little bit of coding,” Senior Vice President of Applications Development Natalia Rachelson told SiliconANGLE in an interview. “These are no-code to low-code people, but we’re now going to go after what’s called pro-code people.”

Oracle Fusion Cloud Applications act as an integrated suite of enterprise software tools that span financial management, human resources, enterprise planning, sales, service, marketing, supply chain, logistics and manufacturing. Large organizations use them to manage transactions and workflows for everyday business operations.

Oracle is now opening that same Fusion-native runtime, with AI agent building and deployment capabilities, to professional developers and the coding agents they already use. This effectively positions Fusion not only as a system of record or execution but as a development platform for autonomous enterprise applications.

“In the case of agentic apps, these apps are made up of teams of agents, each one with their own specialty, and you can give them a business objective, and then they reason and decide how to achieve the objective,” added Rachelson.

Bringing AI coding agents into the fold

With this new update, developers can stay within their own preferred environments. It allows them to work with their existing tools, call upon coding agents and build agentic applications deployed within Fusion. Rachelson named Claude Code, Codex and Gemini as examples of coding agents developers could use. The builder is builder-agnostic, meaning developers can bring it into any agentic development environment, including VS Code, Git and command line interface workflows.

The underlying capability allows developers to describe the intended business outcome and load Oracle’s Fusion AI Studio skill into a compatible coding agent. The coding agent can then generate, validate and test the required Fusion artifacts before the developer deploys the resulting application.

The end result is that a professional developer can use all their usual tools to build deeply integrated, highly capable software that runs within Oracle’s agentic environment. “They’re not really a bolt-on, like an extra layer. They operate inside Fusion,” Rachelson said.

This means that the agents and code operate on transactional data sitting inside Fusion, working with real-time, secure operational company data.

Rachelson used renewal management as an example. An organization could tell an agentic application to increase its renewal rate by a specified amount. The application could examine historical and current records, identify opportunities, draft renewal contracts and potentially send them to customers. The distinction is that it would reason and transact against information already in Fusion rather than operate against a detached copy of the data.

She stressed that the agentic runtime maintains a full governance model providing secure access to business objects, transactions, identity, roles, polices and audit information. It also provides full-service auditability so customers can trace decisions, tools and execution.

This extends the governance theme from Oracle’s October 2025 updates, when the company added monitoring, evaluation, telemetry, token tracking and visibility into agentic changes. The new announcement aims to expand the developer audience that can build for this platform.

Image: SiliconANGLE/Microsoft Designer