Web dashboard products — Head of Design

Part of 8 years of product design from Series B to unicorn — the CSP side

As Head of Design I led web dashboard products for Communications Service Providers (CSPs/ISPs): agent tools, network operations, monitoring, and corporate views. Same bar as the consumer app — enterprise clarity at scale. Dense data, role-based views, atomic continuity with consumer and brand — a peer-connected CSP design engine.

Plume started consumer. By 2018 we were shipping products and messaging for ISP partners who needed visibility and control their legacy tools never gave them. Enterprise ops UI is agentic when it surfaces the next action — not a dump of RF charts.

Today's stack for the CSP suite

Today I'd start from the call, not the screen. Claude would cluster support tickets and NOC escalation notes into an opportunity map. RF interference, bad pod placement, and a weak client radio all show up as the same angry call, and the map separates them by what the agent at each role could actually do about them. The outcome is set per role: a Tier 1 agent resolves or routes correctly on the first call, and the NOC sees an incident forming before the call volume does.

Each problem then gets written as a diagnosis flow in plain text: signal, likely causes, evidence, next action, escalation. Those outlines become generative layouts that get argued over before anything is polished. The survivors are rebuilt in Cursor as a React prototype running on replayed, anonymized fleet telemetry, because a dashboard that only works on a clean happy path fails in the first week of an outage. The interactions are specific: a triage queue ranked by likely cause with the evidence one click away, a pod topology view, one recommended next action per role, and an escalation that sends the full context packet so nobody asks the customer to reboot twice.

One atomic component set runs at different densities. Figma variable modes switch from a calm Tier 1 view to a dense NOC wall, and role permissions decide which modules render. Support agents test the prototype against real incident replays before it ships, and specs and decisions live in Linear. That way I partner with Product and Engineering on what a threshold change means before it reaches a CSP. The job is still to bring clarity to ambiguity: one clear next action for the person in that role.

Time Range

2018–2024

UX/UI

Product thinking

User research & testing

Developer strategy

Atomic design systems

Modular thinking

Product branding

CSP/ISP

Web dashboard products — Head of Design frame

To perfect end user experience we must optimize more than the home, but the ISP network at scale

Challenge

In 2016 - 2018, ISPs had very little insight into their own subscribers network health. They knew how much speed was being served, and how much bandwidth consumed — but hardly more than that.

This meant that ISPs only knew about problems when the customer support lines blew up, and often these were WiFi problems unrelated to their service.

Is the problem with interference from Neighbors? Node placement? Poor WiFi chips on a model of devices? Obstructions in the home such as pipes or appliances?

Solution

With Plume, the ISP can now manage and optimize the WiFi just as customers had always mistakenly assumed they were already doing.

This plus, ISPs could see network health all the way down to the final inches of service as it reaches the destination client device.

We designed multiple customer agent portals, web dashboards, and advanced ISP utilities to create a landscape of awareness and control that was previously unknown territory for ISPs.