Live failure modes on the AI Agents track. Watch task success, tool error rate, steps per task, cost per task, and how often a person takes over. A fall in success with a rise in steps means the agent is thrashing.
This lesson assumes you already worked through Shipping an agent.
The idea in practice
Alert on cost and on repeated identical tool calls. Sample traces daily. Tag new failure types and add them to the suite.
A concrete check
goal = {
'track': 'AI Agents',
'lesson': 'Live failure modes',
}
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 dashboard of only 'users who said thanks' hides loops and double actions. 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
Name three metrics and the direction that means 'stop and look'.
Self-check
- Say Live failure modes 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.