Agentic AI is increasingly taking over key tasks in modern platforms – from tool usage to automating complete workflows. This new autonomy brings enormous opportunities, but also opens up completely new attack surfaces. In this session, we will show in a guided live demo how an Agentic AI system can be compromised step by [...]
DevSecOps for AI-native systems – from code to operation. In this session, we'll show how security is embedded in AI applications from the start: beyond code – including models, data, identities, and pipelines. You will learn proven DevSecOps approaches, automated security along the delivery pipeline and best practices for supply chain, open source and [...]
Building robust agent systems requires much more than simply stringing together prompts in a linear fashion. While many projects start with simple chains, productive deployment requires architectures that can reliably handle errors, process asynchronous events, and manage complex states. In this presentation, we take a concise look at modern orchestration approaches for AI agents, [...]
In this hands-on workshop, we will show how AI agents can be integrated into production applications – with a focus on the open source framework Flock and the Microsoft Agent Framework in AI Foundry. After a compact overview of agent architectures (Blackboard, Tools, Memory), we will demonstrate live the structure of simple agent workflows, [...]
Secure Agentic AI, from architecture to operations. This session shows how Agentic AI systems are holistically secured: from trust boundaries and controlled tool use to protected sources of knowledge to monitoring and incident response. We bridge the gap between risks and concrete protective measures and show how they can be established as a DevSecOps [...]