Appointment brings three decades of systems and security leadership to Kodamai as trust and safety become the gating factor for autonomous agents
Kodamai, a pioneer in mathematically rigorous agentic AI infrastructure that gives enterprises grounds to trust autonomous agents with their most consequential work, announced that Mike Spertus has joined the company as Chief Product Officer, where he will lead product and engineering. Spertus joins as that trust becomes urgent, bringing three decades of driving enterprise-class software at Amazon, Symantec, VERITAS, and the ANSI C++ Standards Committee, along with a long-running post teaching Generative AI, Big Data, and C++ at the University of Chicago.
Also Read: CIO Influence Interview with John Elliott, Cybersecurity Author Fellow at Pluralsight
Mike Spertus has joined the company as Chief Product Officer, where he will lead product and engineering.
The appointment comes at a pivotal moment for enterprise AI. Organizations are moving agents out of pilots and into operations where errors carry financial, operational, and safety consequences. In addition, a run of high-profile agent failures has intensified scrutiny of whether autonomous AI agents can be safely governed at enterprise scale.
Kodamai was founded to address precisely this class of problem. Its Kelvingrove agentic operating system is built on a rigorous mathematical foundation — Category Theory, Type Theory, and formal correctness-by-construction — that constrains what an agent can do before it acts, rather than merely attempting to detect or contain bad behavior after the fact. Agent actions are validated against typed interfaces, producing systems that are explainable and auditable by design across cloud, edge, and device.
Spertus’ record of building enterprise-scale systems used by millions reflects a combination that is rare in practice: the product instinct to know what people will actually use, and the engineering depth to deliver them. He will make the safety benefits of Kelvingrove’s mathematical guarantees ever more accessible through products that enterprises can put to work on their most critical trust problems.
“The incidents we’ve seen from leading AI labs this month are not a story about any one company’s diligence,” said Maha Achour, PhD, Co-Founder and Chief Executive Officer of Kodamai. “They’re a story about the entire industry building agentic systems on infrastructure that was never designed to constrain what an agent can do. Mike has spent his career building systems that are correct and reliable by design, from compilers to distributed data platforms handling trillions of security events. That is exactly the discipline enterprise AI needs now, and exactly why we wanted him leading Kodamai’s product and engineering organization.”
Spertus joins Kodamai from Enklu, where he served as Chief Technology Officer, and previously spent three years at Amazon as a Senior Principal Engineer driving architecture for AWS’ Generative Developer organization. Earlier in his career, he was one of seven Fellows, Symantec’s top engineering position, where he served as Chief Architect for the company’s Authoritative Data Lake, a cross-product store of trillions of security logs that was used to help unmask the Russian Dragonfly attack on the energy sector, and as Chief Scientist for Symantec’s Gartner-recognized Managed Security Services business. He founded and led Geodesic Systems, a venture-backed C++ reliability startup later acquired by VERITAS, and has authored more than 50 proposals to the ANSI/ISO C++ Standards Committee, several of which are now part of the C++ language. Spertus has been a Full Adjunct Professor at the University of Chicago since 2008, where he currently teaches Generative AI, Big Data, and C++.
“Every few weeks brings another headline about an AI agent doing something nobody intended, whether through maliciousness or simple misunderstanding” said Spertus. “The industry’s answer so far has largely been more monitoring and more guardrails bolted on after deployment. Kodamai’s approach is fundamentally different: it makes correctness a property of the system itself, provable before an agent ever touches production. Having spent my career on both sides of that divide, building the reliability infrastructure and watching what happens when it’s missing, I don’t think there’s a more important problem in enterprise software right now, and I’m looking forward to building it with this team.”
Spertus’ appointment follows a period of rapid growth for Kodamai, including the deployment of Kelvingrove at First Mills, Saudi Arabia’s market-leading flour milling company, and the addition of AI and semiconductor veteran Pankaj Kedia to its board of directors.
Catch more CIO Insights: How Are CIOs Aligning Technology with Workforce Agility?
[To share your insights with us, please write to psen@itechseries.com ]
