How Figma Uses AI Agents for Security

Figma’s engineering team recently shared how they created AI agents to support the security team in investigating alerts, searching past incidents and checking internal systems. The agents combine large language models with retrieval‑augmented generation to parse alert details, query logs and retrieve relevant historical tickets automatically. As a result, analysts spend less time on manual data gathering and can focus on validating threats, which shortens the overall investigation cycle. This approach shows that augmenting existing security workflows with purpose‑built AI agents can boost productivity without eliminating the need for human judgment. Other SaaS companies can adopt a similar pattern by integrating LLMs with their own telemetry stores to automate routine security tasks.