Speaker
Description
While Artificial Intelligence (AI) is rapidly becoming a core component of modern cybersecurity, its deployment in operational network environments remains a significant challenge. A key example is Intrusion Detection Systems (IDSs) that must process continuous data streams, respond to threats in real time, and adapt to evolving attack techniques, while, at the same time, remaining stable and scalable under variable network load. Turning AI models into reliable, always-on security analytics therefore requires more than accurate algorithms. It also demands architectures designed for continuous operation, elastic scaling, and rapid adaptation.This talk describes the overall pipeline architecture together with the design choices that enable flexible and scalable AI deployment, along with practical lessons learned when transitioning AI-based intrusion detection from experimental setups to a dependable, cloud-native operational environment.
What will the TNC audience take away from your talk?
Attendees will discover why many IDS deployments struggle in production and how a modular Kubernetes pipeline avoids those pitfalls. The talk shares a practical architecture design pattern that separates the workflow into independent stages, enabling targeted scaling of bottlenecks and safe, rapid updates to ML models without redeploying the entire system.
| Are you a first time speaker at TNC? | Yes |
|---|
