Cisco and OpenAI redefine enterprise engineering with Codex
OpenAI Blog · 2026-05-27
Cisco deployed OpenAI's Codex across production engineering workflows, writing over 95% of new AI features, cutting build times by 20%, and saving more than 1,500 engineering hours per month.
Appears in
Extraction
Topics: enterprise-aiagentic-codingsoftware-developmentcodexopenai
Claims
- Codex wrote the majority of Cisco's AI Defense product and nearly every new feature being built across multiple business units.
- Codex CLI delivered a 10-15x increase in defect resolution throughput on large-scale C/C++ codebases.
- Cross-repo build optimization with Codex reduced build times by approximately 20% and saved over 1,500 engineering hours per month.
- Development timelines for AI Defense features dropped from several quarters to weeks after Codex was introduced.
- Cisco's continuous production feedback helped OpenAI accelerate Codex's enterprise readiness in compliance, long-running task management, and pipeline integration.
Key quotes
Features that would have taken several quarters to get into customers' hands dropped to weeks.
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.
Codex has become a meaningful part of how we think about AI-assisted development and operations going forward.