Inside AI: Chapter 5 – How We Built NINA, From Coding Agents to Agentic Cybersecurity AI

This article presents NINA, an orchestrated multi-agent cybersecurity assistant that augments engineers and analysts by building deterministic security workflows, analyzing large operational outputs, diagnosing and correcting workflow failures, and integrating agentic nodes where valuable. It focuses on human-in-the-loop design, domain-informed automation, cross-node correlation, and faster incident diagnosis and remediation rather than reporting any specific security incident.