Workspaces

ParseSphere lets you have conversations with your data. You upload files—spreadsheets, PDFs, documents—and then ask questions in plain English. The AI reads your files, runs queries, and gives you answers. No SQL knowledge required.

But here's the thing: if all your files lived in one giant pile, every question would search through everything. Your HR policies mixed with sales data mixed with customer feedback. That's a mess.

Workspaces solve this. They're project folders that keep related files together and separate from unrelated ones. Each workspace has its own set of documents and its own conversation history.


How It Works

1

Create a workspace

Start a new workspace for your project (e.g., "Q4 Sales Analysis")
2

Upload your files

Spreadsheets become queryable tables, documents get indexed for search
3

Chat with your data

Ask questions, get answers, follow up naturally

The AI only looks at files in the current workspace, so you get relevant answers without noise from unrelated data.


Creating a Workspace

From your dashboard, click Workspaces in the navigation, then hit Create New Workspace.

Workspaces page with create button

You'll see a dialog asking for:

Create workspace dialog

Workspace Name — Required. Pick something you'll recognize later. "Q4 Sales Analysis" beats "Untitled" every time.

Description — Optional but helpful. A sentence about what this workspace contains saves you from opening it just to remember what's inside.


Private vs Shared Workspaces

By default, workspaces are private. Only you can see them, access the files, or read the conversations. Nobody else—not even your team—knows they exist.

When to Keep It Private

  • Personal analysis or experimentation
  • Sensitive data you're not ready to share
  • Draft work before presenting to stakeholders

Shared Workspaces

If you're part of an organization in ParseSphere, you can share workspaces with your team. Open workspace Settings from the sidebar and toggle sharing on.

When you share a workspace, everyone in your organization automatically gets viewer access. They can chat with the data and see files, but they can't upload or delete anything.

If you need someone to help manage the data, you can promote them to editor. Editors can upload new files and delete existing ones—useful when multiple people need to maintain the dataset.

Roles at a glance:

  • Owner (you) — Full control: rename, delete workspace, manage editors
  • Editors — Can upload and delete files
  • Viewers — Everyone else in your org can chat with the workspace

Information

Sharing is all-or-nothing with your organization. You can't give access to specific individuals only—either the workspace is private (just you) or shared (everyone in your org gets viewer access).


What's Inside a Workspace

The workspace uses a sidebar layout. The left sidebar has three collapsible sections:

  • Chats — Your conversations with the AI about this workspace's data
  • Documents — Upload, view, and manage your files
  • Queries — History of SQL queries the AI ran on your behalf

At the top of the sidebar you'll find + New (start a new chat) and Attach (upload files). At the bottom, workspace owners see a Settings link for managing the workspace description, sharing, and members.


Managing Your Workspaces

Finding Workspaces

The workspaces page has three filters:

  • All — Everything you have access to
  • My Workspaces — Workspaces you created
  • Shared with Me — Workspaces others shared with you

Use the search bar to find workspaces by name, or sort by Recently Used to surface what you've been working on.

Renaming and Deleting

Hover over any workspace in the list to see action icons:

  • Rename — Change the workspace name
  • Delete — Remove the workspace permanently (all files and conversations will be lost)

Updating Settings

Click Settings at the bottom of the workspace sidebar to:

  • Edit the workspace description
  • Toggle sharing on/off (organization accounts only)
  • Manage workspace members

Tips for Organizing Workspaces

One project per workspace. Mixing unrelated datasets leads to confused AI responses. If your sales data and customer feedback don't relate to each other, they belong in separate workspaces.

Use clear names. Six months from now, "Customer Churn Analysis - 2025" tells you everything. "Analysis 2" tells you nothing.

Add descriptions. A quick note like "European region only, amounts in EUR" helps both you and your team understand what's inside without opening it.

Archive instead of delete. If you're done with a project but might need it later, just leave the workspace alone. There's no limit on how many you can have.


What's Next?