Sharing, the AI Assistant, and what to learn next

Two steps remain before you present your analysis to others: getting permissions right and delegating repetitive work to AI. This lesson covers both briefly and wraps the Path.

The permission model in one paragraph

Portal grants permissions at the collection level, and resources inside inherit those permissions. There are three roles:

• Reader — Can view and search, but cannot edit. The most commonly granted role.
• Editor — Can create, modify, and delete resources. Granted to teammates sharing the workspace.
• Owner — Manages permissions on the collection itself. Usually one or two people.

The most common mistake when sharing a dashboard is to share only the dashboard and forget about the datasets it references. Without dataset permissions, the widgets render as empty charts. Granting access at the collection level at once is the safest route.

For the details of permission propagation and exceptions (public links, external users, etc.), see the collection permissions page in the user docs.

AI Assistant — every step of this Path in one line

The AI Assistant in the top bar or right panel is the entry point that lets you invoke an analyst's flow in natural language. You can use it like this:

• Find data: "Are there any revenue datasets from last quarter?"
• Build a widget: "Make a bar chart of revenue by category from this dataset"
• Interpret a chart: "What's a hypothesis for the sharp drop in April on this chart?"
• Generate code: When you need a SQL or Python snippet, call it with ⌘I

The Assistant's answer is always a hypothesis that needs verification. Rather than adopting results as-is, building the widget yourself and comparing is the safer route.

Path complete — what to do next

If you made it here, you've personally walked through an analyst's most common flow once. Recommended next steps:

• Workshop: Retail Inventory Intelligence — Run a real domain scenario end-to-end and repeat the same flow at greater depth.
• Tutorial: Quick scenario import — Load a dhub2-examples scenario into your environment with one command.
• Engineer Path (overview) — When you want to push analysis results into automation and repeatable runs.

If you check off every lesson on this Path, it's recorded as complete. The "Continue learning" panel on the home page will recommend your next flow.

Nicely done.