The launch of Cogent Security is not just another startup announcement — it signals a deep inflection point in the way enterprise cybersecurity is adapting to an AI-driven threat landscape. Born inside Greylock’s Edge program — the same incubator that gave early lift to powerhouses like Palo Alto Networks and Abnormal AI — Cogent is staking its ground with an $11 million seed round and a clear ambition: to radically reinvent how vulnerabilities are discovered, prioritized, and neutralized before attackers exploit them.
This ambition is not abstract. With Vineet Edupuganti (ex-Abnormal AI), Geng Sng (ex-Coinbase), and Thanos Baskous (ex-Blackstone) at the helm, Cogent’s leadership brings battle-tested experience from some of the most security-conscious environments. Their cumulative frustrations — and insights — stem from years of wrestling with the painful inefficiencies of legacy vulnerability management systems: fragmented workflows, sluggish risk assessments, and over-reliance on human coordination that simply can’t keep pace with how fast threat actors now operate using generative AI.
Cogent’s solution is a bold departure from patchwork platforms. Its cornerstone is the development of the world’s first AI taskforce for vulnerability management — a network of specialized AI agents, deeply integrated into the fabric of a customer’s systems and engineered for autonomous action. These agents act like expert-level staff embedded in the organization, ingesting petabyte-scale historical data to understand workflows, infrastructure, and business logic from day one. They don’t just report issues — they reason, prioritize, identify remediations, follow through to implementation, and wrap it all into concise executive-level reporting. All with minimal human oversight.
This automation isn’t generic. It’s targeted. Each AI agent is designed to reflect the decision-making capabilities of seasoned professionals from security, IT, and engineering teams. By turning tasks that used to take days into operations completed in minutes, Cogent is not merely streamlining a process — it’s rewriting the tempo of cyber defense. Initial deployments in Fortune 500 enterprises and major universities already show dramatic gains: fewer manual tasks, faster patch timelines, and measurable reduction in real-world exposure.
Feedback from early adopters underscores the magnitude of the shift. Lucas Moody, CISO at Alteryx, described Cogent not as another dashboard, but as a true “system of intelligence” — a platform capable of operating with the knowledge, intuition, and actionability of an elite human team, working tirelessly across the vulnerability landscape.
This is no small technical feat. Behind Cogent’s sleek AI deployment lies a handpicked team of engineering talent from Tesla, Google Research, and beyond, with deep expertise in building resilient, domain-specific AI systems at scale. With the fresh capital, Cogent plans to grow this team and deepen its investment in custom language models and infrastructure tailored specifically to security use cases.
For Greylock’s Saam Motamedi and Corinne Riley, who led the investment, Cogent represents the perfect convergence of urgent market demand, technological possibility, and rare team competence. Vulnerability exploitation is now the fastest-rising cause of data breaches, and the broader cybersecurity industry is under pressure to reinvent itself in the face of AI-powered attacks. Cogent isn’t trying to iterate on the old paradigm. It’s replacing it — with something faster, smarter, and built for the new rules of the game.
Leave a Reply