The lifecycle of a thunk typically follows three stages: plan, prototype, and run in production. Now that you've built your first thunk, you have a prototype, and the next step is refining it—testing with representative data to ensure your AI agent automation delivers the intended results before moving to production. Often, you might discover that you want to add new steps to the workflow, and refine the existing steps. You will likely need to collect sample data, and repeatedly run it through the workflow while you tune and refine instructions and options to achieve high-quality results.
Step-by-step Guided Demo
In this interactive demo, you’ll see how your thunk runs in the Thunk.AI runtime environment—and learn how to refine it by iterating on steps, grounding the AI with content, and controlling prompts, inputs and outputs as the workflow executes.
Pro Tips for Thunk Iteration
Start small, then scale. Test your thunk with a small, representative data set first. Once you’re confident in the workflow and AI outcomes, scale up to validate reliability across larger volumes.
Instruction changes affect the whole thunk. Any change to a step’s instructions updates the thunk’s plan and applies to all future workflow instances. This is essential because thunks are designed for repeatable work and consistent results.
Iterate on the workflow itself when needed. Sometimes iteration means changing the workflow, not just a step. Use the Workflow Plan Details page to add or remove steps as your process evolves.
Note: Thunk versioning is coming soon. This feature will allow you to create and manage multiple iterations of a thunk. Until then, any changes you make apply immediately to the active thunk and affect all future workflow runs.