Agentic AI for Automated Cloud Vulnerability Management

Maze is an AI-native security company, founded in 2024, that uses AI agents to secure code and cloud.

The Maze platform investigates every vulnerability in a customer’s environment, determines which ones can actually be exploited, and helps teams fix the ones that matter.

The platform was built from inception to think like a security engineer, not with pre-defined rules.

Its agents understand code and cloud together and are trained on millions of real investigations, covering cloud vulnerabilities in containers and VMs, open-source dependencies with AI-SCA, and business-logic flaws in first-party code with its AI-SAST.

Maze customers see ~90% of vulnerabilities turn out to be not exploitable. For the risky vulnerabilities that remain, agents trace each to its root cause, verify the fix, and route it to the right engineer.

Maze has raised $31 million to date. Its $25 million Series A, announced in June 2025, was led by Theory Ventures with participation from existing investors Cherry Ventures and Tapestry VC. The Series A followed a $6 million seed round led by Cherry Ventures.