Run Coworker in our cloud or yours
Start on Coworker's SOC 2 Type II cloud, or deploy the self-hosted appliance in your own AWS or Google Cloud VPC with your models and SSO. Either way you get the same OM2 engine, connectors and permissions, so you can pick what your security review requires.
Coworker-hosted or self-hosted
| Coworker-hosted | Self-hosted | |
|---|---|---|
| Where it runs | Coworker's cloud | Your AWS or Google Cloud account |
| Who runs it | Coworker | Your team |
| Models | Managed by Coworker | Your own models and GPUs, or a provider you choose |
| Security | SOC 2 Type II audited, with annual third-party penetration testing | Inside your own security perimeter |
| Best for | The fastest path to a proof of concept | Security reviews that require data to stay in your environment |
Coworker, inside your security perimeter
Your data stays in your cloud
Your own models and sign-in
The same Coworker your team already uses
Live in your cloud in three steps
1. We deliver it
2. You deploy it in your cloud
3. Your team uses Coworker as usual
FAQ
Frequently asked questions
No. The self-hosted appliance only talks to the endpoints you configure. It makes no calls back to Coworker.
Yes. Run open-source models on your own GPUs, or connect a model provider you choose, such as Anthropic or Google Vertex AI.
Yes. Pre-built connectors for tools like Slack, Google Drive, Jira, GitHub and Salesforce come with it, and each one keeps the permissions from its source. You can bring in custom data, with its permissions, through the API.
Yes. Start with a proof of concept on Coworker's cloud or on sample data while your security team reviews the full deployment.
Install takes days once your data is connected. Your custom knowledge graph is built in about 1 to 2 weeks.
Need Coworker in your cloud?
Tell us what your security review requires and we'll walk through both options on a call.