iNOSnet · Enterprise Network Management · B2B SaaS
From "here's all our data" to "here's what to do about it."
iNOSnet builds KAATHAM Intelligence — an enterprise platform that turns network telemetry into predictive operational intelligence for NOC teams. The product wasn't short on information; it was drowning in it, organized around engineering's internal modules rather than what an operator is actually trying to do. The redesign restructured that information architecture around three operational jobs — Monitor, Investigate, Optimize — alongside a design system, the landing page, and the WordPress build.
- Engagement
- Design sprint — product redesign, plus landing page, WordPress build & hosting
- Duration
- ~2 months
- Scope
- Primary: redesign of the KAATHAM network operations platform (alert triage, GNMI/SNMP analytics, predictions, topology, config management). Plus landing page design, WordPress development, and Hostinger deployment.
- Year
- 2026
iNOSnet's KAATHAM platform is an enterprise network operations tool — the kind of system a NOC team watches all day to catch problems before customers do. It wasn't short on information. If anything it had too much: roughly fifteen modules, fifty-plus leaf-level items, incident screens showing forty-plus fields at once. The problem wasn't missing data. It was that the data was organized around how engineering built the product, not around what an operator is actually trying to do.
The existing navigation mixed four different organizing logics in one menu — a process ("Problem Management"), a data type ("Alerts"), an asset ("Service Routers"), and a UI format ("Dashboard"). To find anything, a user first had to understand the product team's internal taxonomy. The highest-value managerial insights — what's failing repeatedly, where to focus first — were buried three or four clicks deep under a module nobody would think to open.
The redesign's job was to shift the product's whole information model: from "here's every module and metric we have" toward "here's what's happening, why, and what to do about it." Alongside the platform work came a design system, the marketing landing page, and the WordPress build and hosting. Roughly two months, delivered in 2026.
A close audit of the existing product — its structure, labels, and navigation paths — surfaced a consistent pattern: the interface was built around how the system is engineered, not around how an operator thinks during an incident.
Based on a structured IA audit and heuristic evaluation of the existing product, not user interviews or usability testing. The findings are design analysis; validating them with real NOC users was named as the necessary next step.
The navigation mixed four incompatible logics
The top-level menu combined a process (Problem Management), a data type (Alerts), an asset (Service Routers), and a UI format (Dashboard). Users had to learn the product team's internal taxonomy before they could find anything — the navigation reflected the org chart, not the operator's mental model.
The most valuable insights were the hardest to reach
Management-critical views — repeat failures, router hotspots, what to fix first — sat buried three to four levels deep under "Problem Management," a module a manager asking "what needs my attention?" would never think to open.
Everything was a silo; nothing was connected
A router, an alert, and an incident are deeply related in reality — one causes the next. But the product treated each as an isolated page, so answering a single real question ("why does this router keep failing?") meant bouncing between separate modules and losing context each time.
Incident detail showed everything at once
The existing incident view stacked seven-plus panels and forty-plus fields simultaneously — summary, timeline, history, replay, correlations, attachments — with no hierarchy. More visible information, less actual clarity.
The redesigned platform is live and in pilot — iNOSnet is rolling it out with early client companies as the product establishes itself in the market.
The core outcome isn't a prettier dashboard. It's a different information model: the product moved from four mismatched modules to three operator-facing jobs — Monitor, Investigate, Optimize — with a persistent Command Center where the metrics themselves became navigation, and a unified incident view that reveals detail progressively instead of stacking forty fields at once.
It shipped into pilot rather than a full public launch, so what's real to report is adoption in progress, not measured usage gains yet. The improvement targets from the IA audit were projections to design toward, not outcomes measured after the fact — the honest result at this stage is a live product being implemented across early client companies.
No active design work right now, but the relationship is open. The redesign's whole structure — Monitor, Investigate, Optimize — is a well-reasoned hypothesis about how NOC operators think, not something tested with them yet. Validating that with real engineers and managers is the natural next chapter, and the strongest thing that could happen to this work.
Built on solid reasoning. Ready to be proven on real operators.