Cisco and OpenAI Codex Enterprise Integration

Cisco has integrated OpenAI's Codex directly into its production engineering workflows to transform it from a developer productivity tool into an AI engineering teammate capable of operating at enterprise scale. This integration has compressed critical engineering timelines from several quarters to weeks and enabled the automation of complex, multi-repository tasks across C/C++ codebases.

Accelerating Product Development with AI Defense

Cisco utilized Codex to write the majority of its AI Defense solution—an end-to-end AI security product designed to protect against risks introduced by AI—and nearly every new feature currently under development. This shift in development speed has reduced the time required to bring features to customers from several quarters to just a few weeks.

Additionally, Cisco used Codex to develop "Defense Squad," an open-source tool that moved from the ideation phase to the developer community in less than one week. As part of the Daybreak initiative, Cisco also governs access to GPT-5.5-Cyber, a specialized model for cyber defenders.

Agentic AI Capabilities in Enterprise Codebases

Unlike standard code completion tools, Codex provides agentic capabilities that allow it to operate within complex, mission-critical environments. Key technical capabilities include:

• Multi-Repository Reasoning: The ability to understand and reason across large, interconnected repositories.
• Language Fluency: High proficiency in complex languages, specifically C/C++.
• Autonomous Workflows: The execution of CLI-based, autonomous compile-test-fix loops.
• Governance Integration: The ability to operate within existing enterprise security, review, and governance frameworks.

Impact on Critical Engineering Workflows

Cisco applied Codex to three primary high-impact engineering challenges:

Cross-Repository Build Optimization

Codex analyzed dependency graphs and build logs across more than 15 interconnected repositories to identify inefficiencies. This resulted in a ~20% reduction in build times and a saving of over 1,500 engineering hours per month across global environments.

Large-Scale Defect Remediation (CodeWatch)

Through the use of Codex-CLI, Cisco automated defect repair using iterative, agentic execution on large-scale C/C++ codebases. This process increased defect resolution throughput by 10-15x, reducing the time required for remediation from weeks of manual effort to a matter of hours.

Framework Migrations

For the Splunk teams at Cisco, Codex autonomously handled the bulk of repetitive changes required to migrate multiple UIs from React 18 to 19, compressing the timeline from weeks to days.

Enterprise Readiness and Roadmap

Cisco's partnership with OpenAI served as a feedback loop to accelerate Codex's readiness for large-scale enterprise use. Cisco provided critical data on compliance, integration with development pipelines, and the management of long-running tasks.

According to Ryan Brady, a Principal Engineer in Cisco's Splunk group, the most significant gains occurred when the tool was treated as a team member:

"The biggest gains came when we stopped thinking about Codex as a tool and started treating it as part of the team. We use Codex to generate and follow a plan document, allowing the reviewing team to more easily understand both the process and the code generated."