Cisco Principal Engineer's Fix for AI Code Security

About this episode

John Groetzinger discusses CodeGuard, a security skills layer for AI coding agents that improves code security.

Your AI coding agent learned from millions of lines of code, including insecure ones. That means by default, it can write vulnerable code too. So how do you fix that? John Groetzinger, Principal Engineer at Cisco, built CodeGuard, a security skills layer that teaches coding agents how to write and review code securely. He tested it against real scenarios. The result: 84% success rate vs 47% baseline. Nearly 2× improvement. In this episode we get into: how CodeGuard workswhy Cisco open sou...

People in this episode

Host: Tessl

Guest: John Groetzinger

Topics covered

• AI code security
• vulnerable code
• CodeGuard
• software development
• Cisco

Keywords

• AI coding agent
• code security
• vulnerable code
• CodeGuard
• Cisco
• software development
• success rate

Mentioned in this episode

Organizations: Cisco

Products: CodeGuard

More episodes of The AI Native Dev - from Copilot today to AI Native Software Development tomorrow

• Ryan Lopopolo: OpenAI's Framework for Shipping Code at 70 PRs/Week · June 9, 2026 · 56 min
• Why Developers Hit a Wall at 4 AI Agents · June 2, 2026 · 48 min
• Don't Secure the Code. Secure the Coder. · May 26, 2026 · 41 min
• The Hidden Security Risks of AI Coding Agents · May 19, 2026 · 42 min
• "AI Doesn't Stand for Artificial Intelligence" — Venkat Subramaniam's Take Will Change How You Think About It · May 12, 2026 · 53 min
• The Creator of Spring Thinks You Can't Code Serious Software With AI · May 5, 2026 · 57 min