Sonar Summit 2026 | The evolved SDLC: How Cisco scales quality for thousands of developers

At Sonar Summit 2026, Steven Burns, Distinguished Engineer of AI First Engineering at Cisco, shared insights into how the company maintains rigorous quality standards across its massive engineering organization. Burns, who joined the engineering productivity team approximately 18 months ago, discussed his approach to managing quality assurance and code excellence at scale. His role oversees three specialized teams within the Engineering Productivity division, each focused on different aspects of maintaining code quality and developer experience across Cisco's thousands of engineers.

Structuring Quality Engineering at Scale

Cisco's approach to quality engineering is built on three foundational teams. The Quality Engineering team focuses on both testing maturity and code quality across the organization, leveraging SonarQube as a central tool for establishing quality benchmarks. Rather than simply assessing code through traditional QA metrics, the team helps define what quality means at Cisco IT through collaboration with technical leaders. The Engineering Data and Measurement team addresses a common challenge in large organizations: consolidating data from numerous tools into accessible dashboards and insights for engineering leaders. By centralizing SonarQube data alongside other engineering metrics, leaders gain visibility into code quality trends and can drive meaningful conversations with their teams. The Engineering Experience team, functioning as a developer experience (DX) group, identifies friction points in the development workflow and ensures teams have proper access to tools and resources, including SonarQube plugins and custom agents like Koda.

From Engineering Productivity to AI First Engineering

The landscape at Cisco shifted dramatically as AI capabilities evolved. What began as a traditional engineering productivity initiative rebranded into "AI First Engineering," reflecting the strategic importance of artificial intelligence throughout the development lifecycle. Rather than abandoning existing practices, the team integrated AI-first principles into their current operations while maintaining foundational quality standards. Burns established multiple community channels to support this transformation, including a Confluence presence with playbooks and best practices, a Webex space that has organically grown to approximately 4,000 members across Cisco IT and product engineering organizations, and a monthly guild attracting around 500 participants monthly. These initiatives create a unified community where engineers can share experiences, learn from both external vendors and internal successes, and understand that AI integration is achievable within their organization.

Building Momentum Through Community

The enthusiastic adoption of these new initiatives stems from both top-down strategic alignment and bottom-up engineer enthusiasm. By positioning itself in the middle of this ecosystem, the engineering productivity team connects engineers with appropriate tools while removing blockers to adoption. The guild structure, modeled after open source community patterns, creates a rare space within a large enterprise where engineers from different departments can collaborate, share knowledge, and inspire innovation. This organic growth—the Webex space growing at approximately 50 members per week—demonstrates genuine interest in both AI engineering practices and community-driven learning. The combination of structured guidance, accessible tools like SonarQube, and community spaces enables Cisco to maintain quality standards while accelerating adoption of new methodologies.

Key Takeaways

• Three-Pillar Approach: Cisco's engineering productivity strategy balances quality engineering (using SonarQube for code quality benchmarking), data and measurement (centralizing insights from multiple tools), and developer experience (removing friction and enabling access to tools)
• AI-First Without Abandoning Fundamentals: The shift to "AI First Engineering" built upon existing quality practices rather than replacing them, ensuring human-written and AI-generated code both meet rigorous security and reliability standards
• Community-Driven Adoption: Organic growth of guilds and collaborative spaces to 4,000+ members demonstrates that engineers embrace new practices when they have peer support and clear internal examples of success
• Centralized Data Visibility: Consolidating engineering data from tools like SonarQube into unified dashboards enables leaders to have data-driven conversations about code quality across thousands of developers
• Scalable Quality Infrastructure: A dedicated team structure ensures quality standards remain consistent across a massive engineering organization while preparing for emerging practices like agentic and AI-native engineering