WWT Homepage

REGISTRATION FOR THIS EVENT IS CLOSED. IF YOU ARE INTERESTED IN PARTICIPATING, PLEASE MESSAGE Daniel.hartnagel@wwt.com. Your organization is deploying AI. Has your security team ever watched an AI application get attacked, and shown they're prepared to defend it? This live Capture the Flag event puts your team inside a realistic adversarial AI scenario, defending an enterprise RAG application with the same tools securing production AI today. Featured in a workshop at DEF CON's Red Team Village, and now available virtually through the WWT Cyber Range. Tools utilized in the game: 🔷 Cisco AI Defense: monitors and secures LLM-powered applications, detecting prompt injection, data leakage, model misalignment, and unsafe tool/MCP calls 🔷 Isovalent Cilium & Hubble: Kubernetes-native network visibility and Layer 3/4/7 policy enforcement between workloads 🔷 Isovalent Tetragon: runtime process and file-level protection inside Kubernetes pods

Join us for a hands-on look at Check Point's SD-WAN solution, which centralizes routing decisions across multiple WAN links and automatically steers traffic to the best-performing path in real time. In this workshop, you'll see how Check Point builds encrypted tunnels that connect branches, data centers, and cloud networks under a single, centrally managed security policy — improving application performance, increasing resiliency through link diversity, and replacing manual, device-by-device configuration with policy-driven routing intent managed through SmartConsole.

Join us for a hands-on look at Check Point's SD-WAN solution, which centralizes routing decisions across multiple WAN links and automatically steers traffic to the best-performing path in real time. In this workshop, you'll see how Check Point builds encrypted tunnels that connect branches, data centers, and cloud networks under a single, centrally managed security policy — improving application performance, increasing resiliency through link diversity, and replacing manual, device-by-device configuration with policy-driven routing intent managed through SmartConsole.

Every LLM call comes with a cost — in dollars, in milliseconds, and in rate-limit headroom. Most applications default to a single model for everything, and that works until it doesn't: costs spike, latency climbs on simple requests, and a single model outage can take down the whole application. In this session, we'll introduce LLM routing — turning model selection from a hard-coded choice into a runtime decision, so a quick lookup goes to a general-purpose model, a deep analytical question goes to a reasoning model, and a debugging request goes to a coding model, with automatic retries across tiers if one fails. Join us for a hands-on walkthrough of the concepts and a live look at configuring a router to put these ideas into practice.