Speak at The Fifth Elephant 2026 Annual Conference

Modern enterprise networks generate enormous volumes of telemetry, logs, flow records, and packet captures — yet understanding how traffic actually moves through the infrastructure remains surprisingly hard. Network behavior is hidden behind layers of virtualization, overlays, and cloud abstractions, leaving engineers to manually stitch together signals from multiple tools just to answer basic questions about connectivity and traffic flow. This talk explores how we’re using AI on the Nutanix networking layer to change that — automatically discovering topology, understanding how components relate to each other, and building a living model of the network that updates as infrastructure changes.
The core idea is representing the network as a knowledge graph — nodes are devices, services, and interfaces; edges are the actual relationships and traffic paths between them. Once the network is modeled this way, AI can do something useful with it: trace how a packet travels from a VM through virtual switches, overlays, and physical links; correlate telemetry from multiple layers into a coherent picture; and surface insights that would otherwise take an experienced engineer significant time to piece together. The goal is an intelligent foundation — not just better dashboards, but a network that can explain itself.

Key Takeaways:

1. Why graph-based topology representation is the right primitive for building AI-powered network intelligence.
2. How combining topology with flow telemetry and packet traces enables a level of observability that isolated monitoring tools can’t achieve.
3. How this approach transforms network observability — making it possible to get intelligent traces across infrastructure layers, and actually understand what the network is doing rather than just seeing that something is wrong.

Intended Audience

Network Engineers, SREs, Platform Engineers, and Infrastructure Architects who work with complex virtualized or hybrid environments and want to understand how AI can make networks more observable and self-explanatory.

About the Speaker:

Swapnil Singh is an engineer at Nutanix on the Flow Virtual Networking team, where he works on building the virtual networking layer on top of the Nutanix stack.

{Add the link to draft slides - PDF/PPT - with comments access}

{Add the link to 2-min elevator pitch video}