# Multi-Tenancy

> You're reading the Mindset v3 documentation. v3 is no longer sold; this is reference for existing customers. Mindset 4 docs are at /docs.

Deeply personalized AI experiences with secure, tenant-specific agents and knowledge contexts

At Mindset AI, we support **multi-tenant systems** and enable you to create personalized agent experiences using our [**Agent Sessions API**.](https://docs4.mindset.ai/v3/deploy/api/agentsessions-api)

Our [Agent Sessions API](https://docs4.mindset.ai/v3/deploy/api/agentsessions-api) makes it possible for one AI agent to serve many organizations while ensuring each one gets a fully customized and secure experience.

[Video](https://storage.googleapis.com/docs-mindset-ai-videos/Multitenancy.mp4)

## What is an Agent Session?

Each session is created with one **user ID** and one **agent**, but it can include:

- Up to **30 knowledge contexts**
- Up to **5 MCP servers** with workflows inside them
- Up to **10 tags** with metadata (such as tenant IDs or user privileges)

You can think of an agent session as a **secure container** that ties everything together for a single user while they interact with an agent.

Mindset AI provides the flexibility, and you decide how these elements come together in your infrastructure.

![Multi-Tenancy](https://docs4.mindset.ai/v3/assets/images/Multi-tenancy-new-1.png)

## Key Features

### Programmatic Agent Assignment

- Assign the same AI agent or different AI agents that your teams have built to different users (e.g., **HR Agent, Leadership Agent, Technical Agent**)
- Users only see agents relevant to their role and permissions
- Pre-configure agent behavior, tone, and capabilities
- Mix and match any agent with any knowledge base programmatically

### Private Knowledge Contexts

- Each tenant gets their own private knowledge context
- Granular control – users only see knowledge relevant to their role
- Tenants can upload **documents, PDFs, and content** into your platform
- Your team creates knowledge contexts based on tenant uploads
- Real-time content processing and indexing for users

### Flexible Tool Integration

- Different users get access to different workflows based on their role
- Seamless integration with **CRM, HR systems, and monitoring tools**
- Custom workflows built for specific tenant needs

### Session-Based Security

- Every user interaction creates a secure, temporary **agent session**
- Sessions bind together the right **agent, knowledge, and MCP server**, which contains workflows for that specific user

### The Power of Flexibility and Security

Using [the **Agent Sessions API**](https://docs4.mindset.ai/v3/deploy/api/agentsessions-api), you can:

- Package different workflows inside your MCP server and assign them to an agent
- Assign different private knowledge contexts to the same agent so its answers are scoped only to that tenant's material

This gives you **complete flexibility and security**. A single agent can look and feel the same across tenants, but pull from different knowledge and workflows.

## Real-World Example: Corporate Training Platform

Imagine running a **corporate training platform** where different companies use the same AI agent but need access to their own policies and workflows.

You configure a single **Training Agent**. The tenant-specific experience comes from how you create `agentSessions` in your own code.

![Multi-Tenancy](https://docs4.mindset.ai/v3/assets/images/Multi-tenancy-new-2.png)

### Company 1: Cyber Security Company Setup

**Product Manager** gets an `agentSessionUid` that binds:

- Training Agent's `agentUid`
- **Context Z** (contains HR Policy Z)
- **MCP server** with coaching + assessment workflows
- **Tags:** `Company B`, `Manager`

This Product Manager can only access Company B’s HR policies and has management-level privileges.

### Company 2: Logistics Company Setup

**Employee 1: Intern**

- `agentSessionUid` binds Training Agent to **Context A (HR Policy A)**
- **Tags:** `Company A`

The intern only accesses Company A’s HR policies through Context A. No workflows or tools are available.

**Employee 2: CTO (Premium Package)**

- `agentSessionUid` binds Training Agent to multiple contexts: **A, B, C, D**
- Two MCP servers:
  - Coaching + assessment workflows
  - Probation assessment workflow (**premium feature**)
- **Tags:** `Company A`, `Manager`, `System Admin`

The CTO has the **richest experience**, with access to HR, technical docs, financial reports, and all premium workflows.

## Key Benefits

In every case, it’s the **same Training Agent**, but because you choose which `contextUids`, workflows, and tags to attach when creating each `agentSessionUid`, every user gets a **tailored experience**.

Each tenant gets:

- **Private knowledge contexts**
- **Custom workflows**
- **Secure multi-tenant separation**

## Use Cases for Your End Users

- **Role-Based AI Access:** Sales team gets CRM-connected agents, HR gets policy-focused agents
- **Department-Specific Knowledge:** Engineers see technical docs, executives see financial reports
- **Secure Multi-Team Deployment:** One AI platform serves many tenants with complete data separation

## Conclusion

Mindset AI empowers you to turn traditional AI into **flexible, role-aware agents** that deliver real value in contexts like education, HR, or any industry where **personalization and data security** are critical.

Multi-tenancy transforms generic AI into personalized assistants that:

- Understand context
- Remember preferences
- Adapt to each user’s workflow

All while maintaining **enterprise security** and **easy management**.