AI Network Monitoring
GNV@AI
The AI-native evolution of Glean N-View. Ask your network questions in plain language, and get analysis, root-cause findings, and decoded packets back in seconds.
Proven foundation
Built on Glean N-View
GNV@AI carries forward everything Glean N-View does: real-time monitoring, dozens of predefined KPIs, automatic topology discovery, baseline and threshold alarming, and full packet capture. On top of that same trusted data, it adds an AI layer.
Ask your network questions in plain language
GNV@AI puts a conversational assistant on top of your live network telemetry. Instead of clicking through dashboards, you ask, and it navigates to the right data, runs the analysis, and explains what it found.
Ask
Type a question in plain language: "show me host 10.0.0.5 on the core link over the last hour," or "why is this application slow?"
It works the data
The assistant navigates to the right view, pulls the KPIs, decodes packets, and runs the analysis across your topology.
You get answers
Back comes a clear read: what is happening, the likely root cause, and what to do next, with the underlying data one click away.
A platform, not a chatbot
GNV@AI is built in layers, so it fits your environment: your data sources, your workflows, and the model you trust. Everything below the chat box is configurable.
From your question at the top, down to the model that answers it at the bottom.
Six experts behind one chat box
GNV@AI orchestrates a set of specialized agents, each tuned for a different part of the job, and routes your question to the right one.
General agent
Answers questions and guides you through the platform.
Navigation
Takes you straight to the host, link, IP pair, or application you ask about.
Decoding
Decodes captured packets and prepares them for download.
PCAP analysis
Reads packet captures for window sizing, congestion, and anomalies.
Performance analysis
Profiles latency, retransmission, and throughput for hosts, links, and applications.
Troubleshooting
Runs continuously, detects degradation, and produces a root-cause report.
Full analysis reports, in seconds
Ask once; GNV@AI assembles a structured performance report and a root-cause read on the data it just pulled, then suggests what to do about it.
Host performance
Network latency, response time, connection-failure rate, and recommended actions for a single host.
IP-pair performance
Client and server latency, retransmission, and connection state between two endpoints.
Application performance
Response time, retransmission, and congestion analysis for an application.
Link performance
Connection-failure rate, traffic anomalies, and root-cause candidates for a link.
PCAP analysis
TCP window sizing, congestion control, and suspicious-activity detection from a capture.
Troubleshooting
A composite read across connectivity, quality, utilization, and server response.
Run it on the model you trust
GNV@AI is model-agnostic. Connect a hosted API, keep everything inside your network, or plug in the model your organization already standardizes on.
Hosted API
Connect to a hosted model for fast setup and access to the latest frontier models.
Fully on-premises
Run an open-source model inside your own environment. No data leaves your network. Built for closed and regulated networks.
Your own model
Already standardized on an in-house or preferred LLM? Connect it through a standard, OpenAI-compatible interface.
Your packets, KPIs, and topology stay where you put them. Deployed on-premises, GNV@AI runs entirely inside your network with no outbound internet access. It's built for closed and regulated environments such as carriers, finance, and government.
Start with an evaluation
Tell us about your network and what you need to see, and we will scope an evaluation or proof-of-concept with you.