Grafana 🤝 vLLM Madrid: AI Inference, Observability & Security

We are back with a joint meetup with our friends at vLLM Madrid 🤝

Join us at ERNI for an evening exploring AI inference, observability, and security. We’ll look at what makes serving AI models different from running standard microservices, where vLLM makes a difference, and how Grafana can help you spot and alert on exploited vulnerabilities in your cluster.

📍 Venue:
ERNI Consulting España S.L.\, Calle Acanto\, 22. Planta 13\, módulos 3 y 4\, 28045 Madrid | https://maps.app.goo.gl/XyoeRDv2jt6g45h97

🗓 Agenda:
18:20-19:00. Welcome and introductions
19:00-19:30. Grafana VulnObs: See and Alert on Exploited CVEs in Your Cluster
19:30-20:00. AI inference and security
20:00-21:00 Networking with food & drinks

🗣️ Talks
1️⃣ Grafana VulnObs: See and Alert on Exploited CVEs in Your Cluster, Rupesh Koushik
A Grafana plugin that shows the vulnerabilities in every running container image, ranked by real exploit risk using Known Exploited Vulnerabilities (KEV) and the Exploit Prediction Scoring System (EPSS). Get native alerts when an actively exploited CVE appears, right next to your metrics, logs, and traces.
2️⃣ AI inference and security, Nelson Paez — AI Specialist Solution Architect and David Hernandez — SME Technical Team Lead
Why is running inference on an AI model different from running standard microservices, and how can vLLM make a difference? This talk explores those differences and the role of security: why it matters in the AI era and what to consider when working with AI systems.

Thanks to ERNI for hosting and bringing our communities together.

See you there 👋