Main Interface Specification

From Initial Inventory Interview through Complete Dashboard Experience

Status: Draft

Overview

This document defines the end-to-end user experience from the moment a new water system is onboarded through daily dashboard usage for both owners and operators.

1. Initial Inventory Interview (Onboarding)

The system uses a guided, AI-assisted interview to build the initial model of the pumphouse.

Key steps:

Identify major components (pumps, filters, tanks, meters)
Map signals to components
Define healthy vs unhealthy ranges (with AI recommendations + user override)
Set up basic alerts
Configure camera-based sensors if used (OCR or needle gauges)

2. Dashboard Philosophy

Two modes:

Basic Mode (default for owners): 10,000-foot view — simple health status, current key values, active alerts
Advanced Mode (default for operators): Detailed trends, per-component cards, historical data, configuration access

Users can toggle between modes, and their preference is remembered.

3. Monitored Objects

First-class concept. Every pump, filter, tank, or sensor group is a Monitored Object.

Can have relationships to other objects
Has its own visualization card
Has its own alert rules

4. Key Dashboard Sections

Overall system health
Active Alerts
Component cards (pressure, flow, amps, levels, etc.)
Quick links to historical data and alert history

Related: See alerts-spec.md for alert workflows and pumphouse-use-cases.md for persona details.

1.5 KiB Raw Blame History