Straiker, a startup that helps companies keep their AI agents from being hijacked or tricked, raised $64 million in a Series A round announced today. The company is betting that as these agents take on more real work inside businesses, securing them will matter as much as securing any other part of the operation. Straiker was founded by Ankur Shah and Sreenath Kurupati, two people with long histories in cybersecurity who saw agent security as a problem that older tools were never built to handle.
That problem is growing fast. The research firm IDC projects that the number of AI agents actively working inside companies will pass one billion by 2029. That would be a 40-fold increase from 2025. Unlike regular software, these agents can make decisions and take actions on their own, like updating a customer's account or writing code. A single agent that gets tricked into doing the wrong thing can cause real damage, even if nobody steals a password or breaks into a system.
The new round brings Straiker's total funding to $85 million. Marathon Management Partners, Citi Ventures, Illuminate Financial, and Workday Ventures led the round, with continued backing from existing investors Bain Capital Ventures and Lightspeed. Gokul Rajaram, founding partner at Marathon, is joining Straiker's board of directors.
AI agents have spread quickly into everyday business tasks. They now handle everything from simple productivity work to customer service and back-office operations. Because these agents can act on their own, they create a kind of risk that did not really exist before. An attacker does not need to break into a company's systems. They just need to convince an agent to do something it should not.
A recent case shows how serious this can get. This spring, Reuters reported that attackers manipulated Meta's AI support agent into changing account email addresses and bypassing two-factor authentication. That single trick let them hijack more than 20,000 Instagram accounts, and Meta's core systems were never breached. Straiker's own research team, called STAR Labs, found similarly troubling numbers in its testing. 36% of successful attacks on coding agents led to remote code execution, meaning an attacker could run their own commands on a victim's machine. And 91% of attacks on productivity agents led to data quietly leaking out, with no malware and no stolen credentials involved at all.
"Given their scope and autonomy, AI agents need to be treated like first-class citizens," said Gokul Rajaram, Founding Partner at Marathon. "Enterprises need a purpose-built solution given the magnitude of the problem. Straiker will be one of the few companies that define how agents are secured."
Straiker says it will put the money toward three things: building out its product, growing its STAR Labs research team, and expanding into new international markets. According to the company, customer demand has been outpacing its own internal projections, which is part of why it is scaling faster than planned.
The numbers back that up, at least by Straiker's own account. In less than a year, the company says its run-rate revenue, a measure of how much it would earn annually at its current pace, grew more than 15 times over. That kind of growth suggests demand for this type of security is accelerating as more businesses put AI agents into real production use. The funding is meant to help Straiker keep pace without cutting corners on research or product quality.
"Demand is outpacing anything we forecast," said Ankur Shah, Co-Founder and CEO of Straiker. "This round goes straight into product, our STAR Labs threat research, and the global expansion our enterprise customers are pulling us toward."
Straiker was founded in 2025. It describes itself as "The Agentic Security Company." In plain terms, the company helps businesses answer three basic questions about the AI agents running inside their organization: what agents exist, what can they access, and are they behaving the way they are supposed to.
The platform works in three connected steps. First, it helps companies find every AI agent running across their systems, since many organizations do not actually know how many they have. Second, before an agent goes live, Straiker runs tests on it designed to trick or break it, the same way a real attacker might, in order to catch weaknesses early. Third, once an agent is in production, Straiker's runtime protection watches its behavior and steps in to stop threats as they happen. The company says these three pieces work together. A threat caught in production helps sharpen future testing, and a weakness found during testing helps tighten runtime defenses.
Before Straiker, Ankur Shah helped scale Palo Alto Networks' Prisma Cloud business as senior vice president and general manager. Sreenath Kurupati, the company's CTO, led AI and security research at Akamai after that company acquired Cyberfend, a fraud detection business he had founded. Straiker says its work with frontier AI labs gives it early insight into new types of attacks, insight it uses to keep its defenses current as AI systems keep evolving.
"Our uniqueness comes from pairing the industry's most comprehensive agentic exploit dataset with an AI-native security engine purpose-built for autonomous threats. That foundation allows us to secure the next generation of AI-powered enterprises," said Sreenath Kurupati, Co-Founder and CTO of Straiker.
Marathon Management Partners, Citi Ventures, Illuminate Financial, and Workday Ventures led the Series A together. Bain Capital Ventures and Lightspeed, both of which backed Straiker earlier on, returned to participate again. A handful of smaller investors rounded out the round, including GTM Capital, Rain Capital, Neva SGR, Firebolt Ventures, and Sixty Degree Ventures. Together, this group has experience across enterprise software, financial services, and cybersecurity, areas that line up closely with the industries Straiker says it already serves.
This Series A follows an earlier seed round, also backed by Bain Capital Ventures and Lightspeed. With the new $64 million folded in, Straiker's total funding now stands at $85 million. The company says it has built relationships with Fortune 500 companies and frontier AI labs since it launched, and this new funding is meant to support that growing customer base.
"AI agents are getting the keys to real systems and real data, which makes securing them one of the defining challenges of this era. I'm glad to work with innovative teams like Straiker that are focused on getting it right," said Brad Jones, Chief Information Security Officer.
