Operations | Monitoring | ITSM | DevOps | Cloud

Every network team keeps a dashboard backlog. A war room needs a view of one incident. A service owner wants path health for a single application. An executive review needs business impact on one page. Each request becomes a ticket for the tooling team, and the view often arrives after the moment that needed it has passed. Artist is the dashboard agent in Selector Foundry, the agentic NetOps platform built on Selector Analytics.

During an incident, the answers an engineer needs are spread across tools. Telemetry sits in one place, configurations in another, and tickets in a third. Building a complete picture means knowing which tool holds which piece and how to query each one, usually under time pressure. Rosetta is the conversational interface for Selector Foundry, the agentic NetOps platform built on Selector’s existing AIOps and Observability solution.

Network operations teams have heard plenty about AI this year. Most of it answers the alert in front of it, then hands the rest of the incident back to an engineer. Someone still has to correlate the evidence, find the cause, prepare the fix, and prove it held. This week, we announced Selector Foundry, the agentic NetOps solution built into the Selector platform. Foundry adds a team of specialized AI agents on top of the full-stack observability and AIOps foundation our customers already run.

Every network engineer carries an instinct that outsiders mistake for caution: a change in one place can travel. Reroute a path, push a policy, drop an interface, and the effect can ripple across campus, data center, WAN, and cloud before the first alert is read. The blast radius of a network change is the reason operators move deliberately, and it is the single most important thing an AI agent takes on the moment it is allowed to act on the network instead of merely describe it.

It is 2:14 a.m. An agent has correlated a latency spike to an asymmetric routing condition and is ready to reroute traffic away from the affected path. The plan looks right. The only question that matters to the on-call SRE is whether to let it run, and that question is not really about the agent. It is about whether the picture the agent reasoned from is complete enough to trust at 2 a.m. with production on the line.

Most CSP assurance roadmaps now carry an AI line item. Fewer have a clear answer for what that AI actually runs on. Over the past year, the working question across operators and vendors has narrowed to something practical: how to put agents to work in assurance while keeping operators in control.

Every vendor in networking has an agent story right now. The useful question for an operations leader is which of those agents can plan, act, and verify against a trustworthy model of the network, and which are assistants that retrieve and suggest, then leave the decision to a person. The direction of travel is settled.

A reference architecture is a set of design decisions made explicit, and the interesting parts are usually the consequences the authors chose to name. Data center, campus, WAN, cloud, and edge each run on their own platforms, often from different vendors, and the architecture takes that fragmentation as its starting point rather than a problem to wish away.

Walk into a network operations center this year and the useful AI is not running the place. It is doing three specific jobs, and doing them well: cutting an alert storm down to the one incident that matters, pointing at the likely cause, and deciding what deserves a human’s attention first. That is where AI in NetOps pays for itself right now. The part worth noticing is that all three jobs lean on the same thing.

For roughly fifteen years, operating a network has meant living inside a vendor dashboard. An engineer’s skill was, in large part, the ability to read those panels quickly and act on what they showed. Gartner’s read in “The Future of NetOps Is Agentic” is that this arrangement is closing, and sooner than most teams have staffed for.