Before you begin
Make a copy or preserve a known working version before introducing production data or major changes.
Identify the person who will review uncertain, exceptional, or incorrect results.
Prepare the application
AI behavior can change when you update a prompt, model, context source, or output structure.
Before a major change, save
The current prompt
The model and settings
Three to ten representative inputs
The expected outputs
The current eval score
The reason for the next change
Then change one variable.
After the change, run the same examples again.
This gives you a before-and-after comparison instead of relying on memory.
A useful version note sounds like
“Version 4: Added policy exceptions and moved low-confidence cases to review. Kept extraction fields unchanged.”
Do not overwrite a working version because the next idea sounds promising. Preserve the point you can return to.
Common mistakes
Overwriting a working prompt or version.
Failing to record why a version was considered good.
Rolling back code without checking data or configuration changes.
Frequently asked questions
What is the difference between a preview and production?
A preview proves that the workflow can exist. Production adds realistic data, tests, permissions, monitoring, deployment, support, and a defined response when the system is wrong.
Do I need to complete every production step at once?
No. Identify the next material risk and address it. Typical next risks are data, evaluation, edge cases, deployment, or human review.