Escape Raises $18 Million to Scale AI Agents That Automate Pentesting
The YC-backed startup is expanding continuous attacker-simulating agents for live systems to meet enterprise demand.
Overview
- The Series A was led by Balderton Capital with participation from Uncorrelated Ventures, Iris Capital, and Y Combinator, bringing total funding to $23 million.
- Escape’s autonomous agents map apps, APIs, databases, and endpoints, chain exploits with proof of execution, and drive contextual remediation within engineering workflows.
- The company reports more than 2,000 security teams using its platform and over 300,000 assessments per month, with a recent customer citing a 393% ROI.
- Escape’s research found over 2,000 high‑impact vulnerabilities across 5,600 vibe‑coded applications, including 175 instances of exposed personal data in production.
- The funding will deepen agentic pentesting that reasons about application logic and support expansion of engineering and go‑to‑market teams in the US and Europe, with plans to roughly double headcount.