A self-hosted AI SRE you run in your own infrastructure

    Updated 7 October 2026

    Operate runs next to your systems, under your control. Self-hosting decides where Operate runs. Your model choice decides where investigation context goes.

    Deployment

    A Docker Compose stack (web UI and API on port 8080, worker, scheduler, MongoDB, Redis) on any cloud or your own hardware.

    Start from the setup guide or the DigitalOcean walkthrough.

    Permissions

    Adapters read logs, databases, and repositories. Operate cannot commit, open pull requests, merge, or deploy.

    Give each adapter read-only credentials. Some providers ask for extra scopes; the Datadog page lists them.

    Model providers and data flow

    Bring your own model: Anthropic Claude, OpenAI or Azure Codex, an enterprise model, or a self-hosted open-source LLM.

    With a model hosted in your network, investigation data stays there. With an external AI provider, the context an investigation needs is sent to that provider. Operate itself receives only a usage record for billing: App ID, token counts, timings, and model name.

    Auditability

    Every query and agent step is logged, so a reviewer can follow how a conclusion was reached.

    What your team operates

    • The host and Docker Compose stack, including upgrades.
    • Credentials for each connected system.
    • Model provider keys or a self-hosted model.
    • Review and release of any patch.

    Frequently Asked Questions

    Does self-hosting mean no data leaves my network?

    With a model hosted in your network, investigation data stays there. With an external AI provider, the context an investigation needs is sent to that provider. Operate itself receives only a usage record for billing: App ID, token counts, timings, and model name.

    Which cloud providers are supported?

    Any environment that runs Docker Compose and can reach the systems you want investigated, including your own hardware.