RedOps is a Red Labs project — a simulator for practicing the decisions an AI Security & Cyber Defense Lead makes before the real incident forces the issue.
Most AI-security learning is lectures and checklists. RedOps teaches through simulation, consequence, and feedback — you run a security program, make hard calls, and live with the outcomes across a six-month campaign.
AI governance and acceptable use, shadow-AI risk, secure AI adoption, agent and RAG security, SOC monitoring, incident response, vendor risk, and executive communication — the real tradeoffs between security and the business.
Risks and controls map to OWASP LLM Top 10, NIST AI RMF, MITRE ATLAS, and Google SAIF — so the practice transfers to real work.
Next.js, TypeScript, Tailwind, and a rule-based simulation engine. In Live mode, the Anthropic API generates each month’s scenarios from current AI-security news — web-search intel, synthesis, then validation — all server-side.