Editing the config creates a draft, and running commits it as a version. Production runs through the API against a validated version number, while experiments carry on freely on the Studio screen.
Drafts and config versions
Edit a workflow's settings and the next-numbered draft is created automatically; the moment you run, it is committed as a proper config version. Every version is kept, which makes experimenting safe.
How versions are created
- Change a node setting and the dropdown at the top switches to the next-numbered draft. For example, start editing from Config #7 and you're on the Config #8 draft.
- Drafts save automatically in the browser you're using. They aren't visible on other devices or to teammates.
- Press Run and the draft is committed as Config #8, and your cases are processed with it. To commit without running, press Save on the draft item in the config dropdown, or press the Save as v8 button in the Chat tab (Improving with the Chat tab). Applying Quicktune results also creates a new version.
- When a version is committed, a change summary is recorded, such as "Extract step added" or "Mode changed from Standard to Enhanced".
- If you don't need a draft, discard it from the dropdown to go back to the previous version.
What the config dropdown can do
| Feature | What it does |
|---|---|
| Rename | Give a version a meaningful name. For example: "Optimized for FedEx invoices", "Added decimal formatting" |
| Switch version | Select an earlier version and rerun documents against the config at that point in time |
| Compare configs | Turn it on in Table view to compare the results of up to three config versions side by side |
| View step IDs | Check each node's Step ID. Used to identify a node for API integration or support inquiries |
| All configs | Combine the run results of several versions in the Monitor view (Aggregate) |
"Versions stack up by themselves. Experiment without hesitation, and name the meaningful versions so your team can recognize them."
The production rollout process
A workflow running in production through the API is called with a specific, validated config version. Keeping experimentation in Studio separate from production deployment is the core of running this safely.
Start from a baseline
Pin a well-tested config (say Config #12) in your API calls and go live with it.
Keep improving in Studio
Separately from production, keep editing schemas in Studio and building new versions (#13, #14, #15โฆ). Production traffic still uses #12, so this is safe.
Validate the new version
Rerun your reviewed documents against the new version. They are compared against ground truth automatically, which tells you whether the new version is genuinely better.
Ship in batches
Don't deploy every version. Accumulate improvements and roll them out together (every four or five versions, for example). Bump the version number the API points at during a planned window, such as outside business hours.
Roll back if it goes wrong
If a new version performs worse, point the API's version number back; that's the whole rollback. You can also switch back to an earlier version in Studio at any time to reproduce the issue.
Don't over-optimize. A 0.01% accuracy gain isn't worth a production deployment. Wait until meaningful improvements have accumulated across several fields, and use Monitoring to compare versions.
"Production runs on an API call pinned to a version number, experiments live freely on screen. Ship once validated improvements have piled up."
Automating with the API
You can process documents in bulk through the API, with no screen involved. Send documents from your internal systems and load the results into a database for a fully automated pipeline.
Where to find your API details
From the more (โฏ) menu at the top right of the workflow screen, choose Code to open the Use via API panel. It provides a step-by-step reference and code snippets for integrating this workflow into your application.
| Element | What it is |
|---|---|
| Agent ID ยท Config ID | The workflow and config version to call. The Agent ID is the workflow ID (the API still uses the former name, "agent"). Pick the version you want in the Config ID selector (if no config version exists yet, run or Save first) |
| API key | Choose the key to use under Select API key. The default key created at sign-up is connected automatically. Create and manage more keys under Settings โ Organization โ API keys (the Console works too) |
| Code snippets | Step 1 upload a file โ Step 2 create a job โ Step 3 retrieve the job. Toggle between cURL / Python and Linux / Windows commands and copy as is |
| Instructions for AI assistants | The Copy code.md button. Paste these instructions as is when you hand the integration code to an AI coding assistant |
| Go to documentation | Jumps to the detailed API specification (Studio API) |
The call flow
- Upload file: upload the document to process and receive a file ID.
- Create job: request processing with the Agent ID, Config ID, and the uploaded file ID. Because you name a specific config version, your experiments in Studio never affect production (The production rollout process).
- Retrieve job: query status and results by job ID. Each extracted value comes back with a confidence of
low/high. - (Optional) Retrieve reviewed results: for work that needs human review, fetch only the results a reviewer has marked as reviewed and write those to your production database.
Jobs run through the API show up in the Monitoring job list when you set the Source filter to API, and Import to workflow pulls them onto the screen for review.
"The screen is where you build and review, the API is where you run it. An API call pinned to a version number is the core of stable operations."