VeroSecurity, founded in 2025 by CEO Elad Tessler and CTO Daniel Meirov, both veterans of the IDF’s Unit 8200, has announced raising $11 million in a seed funding round led by TLV Partners. The startup is developing a platform dedicated to AI Security Engineering (ASE), meaning securing AI tools and agents, and plans to use the funds to advance its product and strengthen its commercial presence in the United States.
A still poorly defined discipline
AI Security Engineering is a recent security discipline whose contours remain blurry. Its purpose is to secure the AI tools and agents that company employees are increasingly adopting, using, and modifying internally. This phenomenon is accelerating, making the target a moving one, and the resulting requirements differ from traditional cybersecurity.
According to Vero, ASE rules are defined by security teams. The platform transforms these definitions into applied controls, automatically updates these controls as AI tools evolve, and submits only decisions requiring human judgment to the security team.
At the heart of the platform is the AI Runtime Governor, described by the company as a device that enforces security decisions during AI operation. It enforces approved policies and blocks actions by agents that trigger unauthorized behavior. When a change requires a new risk decision, the platform transmits the analysis and recommended action to the responsible security professional.
An open community to train practitioners
Alongside developing its platform, Vero is establishing AISecurityEngineering.org, an open community and resource center offering practical guides, workshops, and an ASE certification path. According to Elad Tessler, cited by SecurityWeek, the weekly number of job postings titled ‘AI Security Engineer’ tracked by the company rose from 367 on October 25 to 451 on November 14, a 23% increase in three weeks. ‘Demand is arriving before the discipline has been formalized, ‘ he said.
The executive notes that other security disciplines, like AppSec or detection engineering, first built a common body of knowledge before having standardized job titles. ‘AI Security Engineering needs that, and right now, ‘ he insists.
For Elad Tessler, the division of labor is clear. ‘The Vero platform handles the impossible part of ASE work, leaving the necessary trade-offs to the human engineer, ‘ he explains. This impossible part relates to volume, with dozens of tool changes each week, each needing to be linked to every control it affects, and to speed, with agents acting in seconds. Defenses must therefore be in place before the action, not examined afterward. ‘The human part is deciding what the company accepts, which exceptions are justified, and when a new tool is worth the risk, ‘ he adds.
‘We are betting on more AI security engineers, not fewer, ‘ asserts Elad Tessler, who presents AISecurityEngineering.org as a means to train them. According to him, without Vero, keeping pace with AI in a large company requires an entire team. With the platform, a single person, or a security manager handling it alongside their main role, can stay ahead.
