AI Workloads
An AI workload is an automation-ready task that requires human judgement and can be meaningfully measured by an expert — processing a batch of documents, producing a report, reconciling records. If you can define what a good result looks like, and the task will come up again, it's a workload Jetty can run. See Running AI Workloads for the full lifecycle.
Every workload is built from two pieces:
- A runbook — the plain-language markdown file that gives the agent its job, its inputs, its bar for “done”, and the checks that enforce it.
- A runtime — the coding agent and model that carry the runbook out inside Jetty's sandbox, swappable run to run without editing the file.
The hero artifact: one markdown file with the job, the bar for “done”, and the checks.
The agents Jetty runs inside its own sandboxes (claude-code, codex, gemini-cli, antigravity, opencode, hermes, pi, goose), which providers back each, and how to pick one.
Provider keys and environment variables, and how sandboxed runs read them.
Every run is captured end to end. Evals turn runs into a hill-climbing loop with /optimize-runbook.
One API, two modes (passthrough proxy and runbook sandbox), durable execution, object storage.