About
Cybersecurity leader with more than 25 years of experience across advanced threats, incident management, SOC architecture, MDR/MSSP operations, EDR, XDR, SOAR, malware analysis, threat intelligence, anti-fraud, and anti-phishing. Career experience includes senior roles such as Lead Architect Cybersecurity – Advanced Threats Group at Cisco, Principal Advisor, SOC Senior Lead, Senior Security Specialist, and Security Consultant, working with organizations including UL – Underwriters Laboratories, ServiceNow, Reply / Communication Valley, and DFLabs. A significant part of this journey has also been dedicated to cybersecurity research, particularly botnets and malware, with more than 20 published research articles and multiple conference speaking engagements. Today, the focus is on Imperum and the next evolution of security operations. The SOC is moving beyond dashboards, alert queues, and manually executed workflows. Agentic AI, autonomous investigation, intelligent automation, and continuous machine reasoning are creating a new operating model for cybersecurity. Imperum is being built around that transition: bringing AI agents, detection, investigation, response, automation, and human expertise together into a unified Autonomous SOC platform.The goal is not simply to make analysts faster. It is to rethink how security operations are designed, how decisions are made, how knowledge is retained, and how organizations defend themselves at machine speed. International experience and written and verbal communication across six languages have also supported work with diverse customers, partners, security teams, and technology organizations across multiple markets.The next chapter of cybersecurity will be defined by the convergence of human expertise, autonomous systems, and Agentic AI. That is the direction Imperum is built for.
Sessions
WHEN AI GOES ROGUE: Jailbroken LLMs, Autonomous Agents, and the New Cyber Threat
What you will learn:
What happens when an LLM stops refusing—and starts acting? This session explores how jailbroken and rogue LLMs, combined with autonomous agents and access to real tools, can transform prompt manipulation into real-world cyber risk. We examine recent incidents, the emerging agentic attack surface, and demonstrate how Imperum Agent Studio can simulate these threats while Virtus Sentinel detects rogue LLMs, prompt injection, jailbreak attempts, and malicious agent behavior—turning the attacker’s AI advantage into a powerful tool for continuously testing and hardening cyber defenses.