Shipping an agent on the AI Agents track. Production means limits, traces, an owner, and a way to turn the agent off. The off switch disables tools, not just the chat box.
This lesson assumes you already worked through Task suites for evals.
The idea in practice
Ship behind a flag. Start with a small percent of traffic or an internal team. Page a human when the error rate or the cost per task jumps.
A concrete check
goal = {
'track': 'AI Agents',
'lesson': 'Shipping an agent',
}
checks = [
'input available at decision time',
'score matches the real decision',
'failure case written down',
]
print(goal['lesson'])
for item in checks:
print('-', item)
Run the sketch locally if you have Python. The printout is a reminder of the checks, not a trained model. Replace the strings with the real inputs from your own example before you treat it as a design.
What usually goes wrong
A launch with no off switch leaves you waiting for a deploy while the agent keeps acting. When this happens, stop adding parameters or tools. Fix the check, the data, or the permission, then run the same example again.
What to write down
- The input you are allowed to use at decision time.
- The output and the score or pass rule.
- One failure you will test on purpose.
- What you will not claim the system can do.
Practice
Write the rollback steps: what you disable first, and how you finish in-flight runs.
Self-check
- Say Shipping an agent in one sentence that mentions an input and an output.
- Name the failure mode in this lesson and the check that would catch it.
Done when: you can explain this lesson without the page open, and you have a written failure case.