Get StartedYour first five minutes on the home dashboard

5. Your first five minutes on the home dashboard

A guided tour of the page Daalu opens to — what each panel means, and what it shows before you’ve connected anything.

The home dashboard is the page Daalu opens to when you log in. It’s at https://ops.daalu.io/ — no path. Spend a few minutes here on your first day and you’ll already know where everything else lives.

This chapter walks the page from top to bottom, panel by panel. Where a panel needs data Daalu doesn’t have yet, we explain what you’d need to connect to populate it.


The layout

The page is divided into three columns:

+---------+----------------------+----------+
| sidebar | main column          | rail     |
| (nav)   |                      |          |
|         |  - greeting hero     |  - AI    |
|         |  - daily briefing    |  - notifs|
|         |  - summary tiles     |  - active|
|         |                      |  - health|
+---------+----------------------+----------+
  • Left sidebar — the global navigation. Every other page in this book is reached from here. Once you’ve logged in once or twice, you’ll know it by muscle memory.
  • Main column — what’s happening in your tenant right now. The full daily briefing sits at the top so you don’t have to click through to read it.
  • Right rail — the AI Assistant, your notification feed, any active automation sessions, and a system-health gauge. We cover the Assistant properly in Chapter 13.

The greeting hero

The top of the main column is a soft, glowing hero panel with:

  • “Good morning, Name.” (Or afternoon / evening based on your local time.)
  • A one-line operational summary — “All systems are running smoothly” or “N items need review” if criticals are open.
  • Subtle ambient particles and the Daalu AI core illustration.

Click anywhere outside an interactive element to leave the hero untouched — it’s a status card, not a button.


The daily briefing

Directly beneath the hero, the main column shows the latest generated daily briefing in full. This is a few paragraphs of plain English that describe what’s happened in your tenant since you last looked.

The briefing is composed by an agent that runs once a day at your configured briefing time (default: 8 AM in your tenant’s time zone). It pulls from:

  • Recently-resolved and currently-firing alerts.
  • New devices discovered by the network reconciler.
  • New cloud resources detected (e.g., “12 new EC2 instances came up in us-east-1 yesterday”).
  • Open change proposals waiting for approval.
  • Anomalies in spend that exceed the dollar threshold set in Settings → Briefings.

The card has a Regenerate button in its top-right corner. Click it to ask the agent to write a fresh briefing from current data — useful when something just happened and you don’t want to wait for tomorrow’s scheduled run.

On a brand-new tenant the briefing will be honest about not having much to say: “No briefing yet for the infra channel. To generate one now, click Regenerate. To feed real data in, configure an integration on the Integrations page or POST events to /api/v1/events.”

Historical briefings are archived on the Reports page (Chapter 23), under the Briefings tab. Click any past briefing in that list to re-read it.


Summary tiles

Beneath the briefing is a grid of small tiles, each showing a single signal and linking through to the relevant detail page:

  • Active agents — count of autonomous helpers currently online. Links to the Agents page.
  • Open incidents — current open incidents with a critical count. Green hint if it’s all clear; red if it isn’t. Links to Operations.
  • Operational activity (24h) — count of events ingested in the last day, with a tiny sparkline showing the 24-hour histogram. Links to Alerts.
  • Workflows running — automation sessions in flight, with how many completed today. Links to Automations.
  • Briefings on file — archived briefing count. Links to Reports.

Each tile is a single click target. Hover to see the link target in the browser status bar.


The right rail

Anchored to the right side of every page (not just home):

AI Assistant

The chat input is always one click away. Type a question and the Assistant responds with grounded answers — citing the metrics, logs, or events it drew on. Three suggested prompts appear when the conversation is empty:

  • “What services are degraded?”
  • “Summarise today’s events.”
  • “Which alerts need attention?”

Conversations persist per user across sessions.

Notifications

A live feed of three groups:

  • Critical — open critical alerts (newest first).
  • Warning — open warnings.
  • AI Generated — pending recommendations the agent has surfaced for review.

Every notification is clickable. A critical or warning item opens the alert’s detail page (e.g., /alerts/c5936f76-…). A recommendation item lands on the Proposals page where you can review and approve or dismiss.

Active sessions

Workflow / automation runs currently in flight. Each row shows the workflow name, when it started, and whether it’s running or waiting for approval. Useful when the AI is doing something on your behalf and you want to know about it.

System health

A small radial gauge at the bottom showing percentage reliability — computed from the count and severity of open alerts. 100 % means nothing is firing.


The sidebar in detail

Quickly, what each item in the left sidebar does. We’ll cover each in its own chapter in Part IV.

  • Home — what you’re on now.
  • Operations — events, incidents, AI insights filterable by module.
  • Agents — autonomous helpers and their run history.
  • Automations — workflows you’ve authored, with live run status.
  • Reports — archived briefings (default tab) and saved analyses / ad-hoc queries.
  • Alerts — open and resolved alerts.
  • Proposals — pending change proposals waiting for human approval.
  • Devices — your network device list (when Source of Truth is connected).
  • Integrations — wire up Slack, Prometheus, Loki, cloud accounts, etc.
  • AI Factory — your connected GPUs: health, utilization, inference routing, diagnostics, and load tests.
  • Workspace — a browser-based, AI-assisted coding environment backed by your own GPU inference.
  • Managed infra — federated customer clusters connected via Daalu Edge.
  • Usage & Pricing — spend and inference cost.
  • Onboarding — guided setup wizard.

At the very bottom of the sidebar:

  • Settings — profile, preferences, API tokens, notification routing.
  • Help & Feedback — version info and a way to write to the team that builds Daalu.

What a populated home page looks like

For comparison, here’s roughly what the page looks like a month into a real customer’s tenant:

  • Greeting: “Good morning, Jane. 2 items need review.”
  • Briefing: three paragraphs. The first summarizes overnight cloud activity (“EC2 instances scaled in by 4; one new RDS snapshot retained from the daily job”). The second covers alerts (“kafka-broker-disk-pressure fired at 03:12 and resolved at 03:18 — agent traced it to a temporary log rotation spike”). The third lists pending decisions (“ChangeProposal #4317 from the SoT reconciler wants approval — a new BGP neighbor on edge-router-3”).
  • Summary tiles: real numbers — 4 active agents, 1 open incident, 1,247 events in the last 24 h, 2 workflows running, 31 archived briefings.
  • Right rail: 1 critical, 2 warnings, all clickable; AI Assistant ready; 1 workflow waiting for approval; 96 % reliability.

On a brand-new tenant the page reads like a quiet morning. That is expected and not a bug — connect a source on the Integrations page and signals will start flowing.


What’s next

Chapter 6 covers inviting your team. Even if you’re not ready to do this on day one, glance at the chapter — there are a few gotchas (role choices, what happens on offboarding) that are useful to know before they bite you.

Chapter 7 walks through connecting your first cloud account, which is what makes the home page actually have something to say tomorrow.

Next: Chapter 6 — Inviting your team