Use case · Products
Multi-tenant agents.
One agent serves all of your customers. Each customer only ever sees its own data.
You run one system instead of one per customer. A forgotten filter can't show one customer's data to another.
How it works
Step by step.
Press play, or click through it yourself. Try the switch above the drawing.
Try it · Ask as
acmeglobexacme dataglobex dataacme userglobex userone agent
01Two companies use your product. Each one has its own data.
01 / 06
Where it's used
One idea, many industries.
- 01SaaS supportA help-desk product answers the end users of every customer company from that company's own tickets.
- 02HR softwareAn assistant answers employees' questions from their own employer's handbook.
- 03Property managementEach landlord's assistant sees its own buildings, tenants and contracts.
- 04Accounting softwareEach firm's assistant works only with that firm's clients and books.
How Promptise makes it work
All three parts, working together.
In code
A peek at the code.
- 01Each request names its user and its company.
- 02Memory, chat history and the cache are stored per company.
- 03The model never picks the company itself.
support.pypython
from promptise import CallerContext # your app knows who is asking; the agent keeps them apartcaller = CallerContext(user_id="maya", tenant_id="acme") result = await agent.ainvoke( {"messages": [{"role": "user", "content": "Where is order 4471?"}]}, caller=caller,)A real project you can run
Customer support
Multi-tenant support agent
One support agent sold to many customers, where each customer's data, memory and limits stay their own.
A reference application with local stand-ins for the outside systems.