Scaling Code Quality at Cisco | The Evolved AI-Driven SDLC
summary
Cisco embeds SonarQube intelligence into an agentic SDLC, running an autonomous teammate called Koda agent in Jira that pulls Sonar issues, creates plans, raises PRs and closes tech debt. Teams feed Sonar context to Copilot and MCP servers, build verification checks into plan→implement loops, and track defect density and quality-gate pass rates. Reported outcomes include a 3× increase in story point delivery, 27,000 legacy issues reduced to near zero in months, and automated pre-PR verification at scale.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 0:56Steven’s Role and Cisco’s Productivity Teams
- 3:52Koda and Reducing Developer Friction
- 4:17Joining Cisco During the AI Transformation
- 5:53Building Guilds for AI-First Engineering
- 8:57From Quality as Table Stakes to Quality as Accelerator
- 10:04Using Sonar and Agents to Tackle Technical Debt
- 12:22Koda as an Autonomous Teammate in Jira and PRs
- 14:53Driving Toward Tech Debt Zero with AI Workflows
- 17:48Measuring Speed, Quality, and Developer Happiness
- 21:42From Copilots to Agentic Engineering Workflows
- 33:10Shared Quality Standards and Verification at Scale
related talks
16mShift Left with GitHub Copilot: Enforcing Code Quality Before the Commit
43mCode quality as an engine of your IT business
36mAgentic CI/CD: from Pull Request to Production without the Paper Cuts
46mCode Quality in the AI era, needed more than ever
06mWith great AI power comes great responsibility
40mSecure by Default: Building an AI-Augmented, OSS-Powered Reusable CI/CD Pipeline
26mOur Journey to Agentification
31mThe Copilot Wrote It but Who Saves It: Surviving the AI Code Tsunami