Skip to content
Guide
Cookbooks

Enterprise Rollout

Configure LlamaCloud for a large enterprise: organization and per-team project structure, shared embedding/data-source/sink integrations, roles, and a phased pilot-to-teams rollout.

Rolling out LlamaCloud in a large enterprise requires planning to ensure a seamless onboarding experience for users. This guide provides step-by-step instructions for configuring LlamaCloud for an enterprise environment, defining user roles, setting up integrations, and best practices for operationalizing LlamaCloud in production.

Note that we will be expanding our RBAC offerings, and this cookbook will evolve with additional features

LlamaCloud structures access and resources using Organizations and Projects.

  • Navigate to Settings
  • Define a single organization for your enterprise (e.g., ACME instead of each user having their own organization).
  • As you scale, you can consider additional organizations. We have seen large enterprises create an organization per business unit.

1.2 Configure Projects for Departments/ Teams

Section titled “1.2 Configure Projects for Departments/ Teams”

Each Project serves as a logical unit within the organization. Recommended structure:

  • Experiments: For initial testing and onboarding
  • Teams: Create dedicated projects per Team (e.g., Research, Engineering).

Note that integrations are scoped to a project (see below).

You can pre-configure integrations to streamline workflows for your users. There are 3 kinds of integrations

  • Embedding Model (e.g. OpenAI keys)
  • Data Sink (e.g. vectorDB like MongoDB)
  • Data Source (e.g. Sharepoint or Box)
  • Embedding model API Key Management: Configure Embedding Model connection as a shared project resource. When creating an Index, a user can select from a dropdown the already configured API key instead of entering credentials manually.
  • Pre-configure Data Source Connectors: Use a service credential to provide controlled access to shared folders.
  • Pre-Configure Data Sink Connection
  • When creating an index, users can select Data Source and Data Sink from a dropdown list

In Settings → Members you can add team members to your organization. By default, LlamaCloud has two roles: Admin and Viewer. We plan to add additional roles for granular control.

  • Start with a small team of 10-25 pilot users in the Experiments project.
  • Gather feedback on usability, security, and role assignments.
  • Roll out LlamaCloud to various teams (or use cases) by creating projects for each.
  • Configure resources for each project so that the Index creation process is seamless
Note for AI agents: this documentation is built for programmatic access. - Overview of all docs: https://developers.llamaindex.ai/llms.txt - Any page is available as raw Markdown by appending index.md to its URL — e.g. https://developers.llamaindex.ai/llamaparse/parse/getting_started/index.md - Agent-friendly REST search APIs live under https://developers.llamaindex.ai/api/ — search (BM25 full-text), grep (regex), read (fetch a page), and list (browse the doc tree). See https://developers.llamaindex.ai/llms.txt for parameters. - A hosted documentation MCP server is available at https://developers.llamaindex.ai/mcp. If you support MCP, you can ask the user to install it for browsing these docs directly (an alternative to the REST API). Setup: https://developers.llamaindex.ai/python/shared/mcp/