The Tel Aviv startup, founded by veterans of McAfee and RSA, is defining a new category it calls agentic cloud security. Its Context Lake gives AI agents the architectural understanding to investigate and remediate threats without human intervention, aimed at reducing reliance on manual workflows that have kept security teams buried in alerts.
Cloud security has a backlog problem that may not be fully addressed by scaling teams alone. As enterprises expand across hundreds of cloud accounts, the volume of alerts, findings, and potential vulnerabilities has, in some cases, exceeded the practical capacity of human security teams to review them. A Fortune 500 company might generate millions of raw security findings in a given period. Triaging those findings, determining which represent actual risk, and executing remediation has remained a largely manual process, even as AI has transformed nearly every other domain of enterprise software. That gap is what Copperhelm was built to close.
Copperhelm, a Tel Aviv-based cybersecurity startup, emerged from stealth today with $7 million in seed funding and what it describes as among the earlier examples in the space of an agentic cloud security platform. The round was led by TLV Partners with participation from toDay Ventures. Shay Michel, Managing Partner at Merlin Ventures, joins the company's board of directors as part of the announcement. The company reports it is already working with paying customers, including Fortune 500 enterprises.
Copperhelm was founded by Shimon Tolts (CEO), Eyar Zilberman (CPO), and Roman Labunsky (CTO), who was previously the VP of Engineering at Datree, a developer-focused Kubernetes security tool. TLV Partners is doubling down on the same team. The founders have a recognized standing in the cloud infrastructure community: Tolts holds AWS Hero and CNCF Ambassador designations, and Zilberman is a GitHub Star. Their prior work at McAfee and RSA gave them firsthand exposure to the operational complexity of large-scale cloud environments, and the specific friction point that became Copperhelm's founding thesis.
"Engineering teams got AI years ago; security was left behind doing manual work. Copperhelm finally brings true AI to cloud security. It's like instantly adding twenty senior engineers to your team." -- Shimon Tolts, CEO and Co-Founder, Copperhelm
The Context Lake: why general-purpose AI fails at cloud security
The founding team's core insight, and the reason they spent significant effort before shipping a product, is that cloud security cannot be solved by connecting a general-purpose AI model to a company's cloud accounts. The challenge is context. A large enterprise cloud environment spans hundreds of accounts, services, workloads, and interdependencies that evolve continuously. Without a deep architectural understanding of how those components relate to each other, an AI agent cannot reliably determine whether a given finding represents a real, exploitable risk or a false positive. And an autonomous agent acting on an incomplete context in a live production environment is a liability, not an asset.
Copperhelm's answer to this is the Context Lake, a proprietary data foundation that structures and connects cloud data across environments in real time. The Context Lake is not a log aggregator or a SIEM replacement. It is a purpose-built layer that gives Copperhelm's AI agents the architectural comprehension required to reason about a complex multi-account environment. According to the company, the system is designed to operate in a way that reflects how experienced cloud engineers might assess vulnerabilities, including how they relate to broader infrastructure and potential remediation paths.
Agents that investigate, validate, and remediate
Built on the Context Lake, Copperhelm deploys specialized AI agents focused on distinct functions: network analysis, system behavior monitoring, adversary simulation, and automated remediation. These agents connect directly to live workloads, inspect active processes and container images, map cloud network topology, and can deploy targeted protections such as WAF rules to mitigate threats, which the company says is intended to minimize disruption. The platform is designed so that human security teams remain in control, with agents handling the investigation and remediation pipeline autonomously, which the company describes as focusing on more validated, evidence-based risks that require human decision-making.
The practical impact of this architecture may be meaningful in certain contexts. The company reports that, in one enterprise deployment, a large volume of security findings was reduced to a significantly smaller set of prioritized risks, enabling teams to focus on a more manageable subset of potential issues, which reflects a broader shift in how security operations can function. The bottleneck is no longer analytical capacity. It is execution capacity, which is a problem that engineering teams know how to solve.
"Applying AI to cloud security requires deep architectural expertise, not just generic models with integrations. Shimon, Eyar, and Roman are true cloud veterans and the right team to bring autonomous AI into this space." -- Rona Segev, Co-Founder and Managing Partner, TLV Partners
A new category: agentic cloud security
Copperhelm is positioning itself within an emerging category it calls agentic cloud security, a term meant to distinguish its approach from the alert-generating security tools that currently dominate the market. The distinction is not semantic. Most cloud security platforms today identify potential risks and surface them for human review. Copperhelm's agents are designed to investigate, validate, and remediate those risks autonomously. The company suggests this approach may shorten the time between identifying and addressing issues compared to more manual triage processes. The company states that the $7 million in seed funding is expected to support accelerated product development, expand go-to-market efforts, and scale the engineering team. The company's stated goal is to reach a point where security analysts never have to manually investigate a cloud alert again, with agents handling 99 percent of threat validation and mitigation autonomously, leaving human teams to focus on architecture, policy, and strategy.
The timing of the launch reflects a broader shift in enterprise security budgets toward autonomous and AI-driven tooling, driven in part by the accelerating pace of adversarial activity. Copperhelm's argument is that the only sustainable response to adversaries using AI to move faster is defenders using AI to respond at the same speed.
Copperhelm is headquartered in Tel Aviv, Israel. The company emerged from stealth on April 23, 2026. More information at copperhelm.com
VentureBeat newsroom and editorial staff were not involved in the creation of this content.