Workflow
Catalyst Workflows are long-running, stateful orchestrations with checkpointing, retries, durable timers, and fan-out/fan-in. They are built on the open-source Dapr Workflow API, using the same authoring model — your existing Dapr workflow code runs on Catalyst unchanged.
Every workflow step is persisted so that when a process restarts, activity failures, or a tool call times out, the workflow resumes from the exact point of failure instead of starting over. This makes workflows the right choice for:
- Long-running business processes — multi-step orchestrations that span seconds to days.
- AI agent execution — durable agents that survive process restarts and LLM call failures.
- Fan-out/fan-in — parallel activity execution with aggregation.
- Durable timers — scheduled steps, polling loops, and human-in-the-loop approvals.
For language-specific guides see Develop workflows. For inspecting running workflows in the console, see Operate Workflows.