Proterra · IoT Leadership

Clarity as a Design Deliverable

Bringing order to a system that nobody fully understood yet

260

Devices managed across the MDC dashboard

Status states — designed for uncertainty, not binary on/off

Products — Transit platform + MDC controller

Role

Design Lead

Company

Innominds for Proterra

Scope

Transit control + MDC platform

Domain

IoT · EV Infrastructure · Distributed Systems

../assets/proterra-hero.png

MDC exception-first dashboard — 260 devices across 4 status states (Charging/Faulted/Offline/Booting) with colour-coded status bar

The Challenge

The hardest problems weren't design problems — they were clarity problems

Proterra's MDC platform managed distributed EV charging infrastructure via IoT and cloud. Requirements were ambiguous, stakeholders were misaligned, and engineers who would use it daily had no mental model of how it all connected. Before any visual design, I had to treat the ambiguity itself as the first design problem to solve.

Stakeholder

Conflicting priorities, no shared definition of done

Different stakeholders had different assumptions about what the system should show, what actions it should allow, and what "control" even meant in a distributed IoT network.

Requirements

Evolving scope in a technically complex domain

What looked like a product decision on day one was often a systems capability question that needed engineering input before a design direction made sense.

User

Engineers without a usable mental model

Technically expert engineers — but the existing workflows didn't match how they actually operated. The system's logic was opaque. Obvious decisions required navigating through several layers of state and interface.

System

Real-time IoT with inherent unpredictability

Distributed dispensers across multiple locations. Real-time status that changes without user action. Commands that may not execute immediately or at all. The UX had to account for states on a spectrum between "working," "uncertain," and "failed."

The turning point

Clarity is a design deliverable — not a prerequisite for design. The documents, frameworks, and shared principles produced during stakeholder alignment were as consequential as any wireframe that followed. Anyone can produce screens. Not everyone can take a technically complex, stakeholder-conflicted product and bring it to a point where the right design becomes obvious.

Key Decisions

The choices that defined the product

Surface alerts before all-device status

Completeness vs relevance — relevance wins.

Stakeholders wanted engineers to "see everything." But engineers didn't want to see everything — they wanted to know what was wrong. The MDC dashboard led with a single-row status bar showing all 260 devices across four states, colour-coded and immediately scannable. Faulted devices visually distinct before any interaction.

Design for uncertain states, not binary states

Four states: Charging, Faulted, Offline, Booting.

IoT systems don't have clean on/off states. A dispenser might be "connecting," "unresponsive," or "last seen 4 minutes ago." Every status indicator was designed to communicate confidence level, not just current value — because in a distributed system, "unknown" is a valid and important state.

Constrain controls to what the system can reliably execute

Fewer actions, each clearly scoped to guaranteed capabilities.

Stakeholders initially wanted broad remote control — start sessions, override states, force resets. Engineering flagged that many commands were unreliable over the IoT layer. Rather than showing controls that might silently fail, I advocated for a narrower, more honest control set. Trust in a monitoring tool depends on it not lying to you.

../assets/proterra-dashboard.png

MDC device detail view — configuration parameters, firmware version, security certificates, and 6 app packages in a single scrollable view

Impact & Reflection

Clarity delivered as a product

Product

Task-centric IA replaced system-centric navigation
Uncertain states designed explicitly — engineers trust the display
Control set constrained to reliably executable actions

Engineering UX

Engineers identify and act on network issues without navigating the full device hierarchy
Real-time status model aligned with how IoT systems actually behave
Reduced cognitive overhead in high-attention operational workflows

Process

Stakeholder alignment delivered before first wireframe — fewer late-stage direction changes
Design principles documented — decisions in later sprints had shared grounding
Demonstrated that design leadership in complex domains starts before the visual work

The leadership principle this project confirmed

In complex systems, the designer's most valuable contribution isn't the design — it's the clarity. In a domain like IoT infrastructure, a designer who doesn't understand the system's actual behaviour will produce interfaces that look reasonable but fail operationally. Building enough technical understanding to design honestly, not just visually, is what separated this work from a surface-level redesign.