Design a complete agent on the AI Agents track. The capstone is one agent you could defend: a written spec, three tools, one approval gate, a ten-case suite, and a trace of a failure.
This lesson assumes you already worked through Interview review.
The idea in practice
Keep the spec to two pages. Include budgets and the off switch. Run at least the failure case on paper if you cannot run code.
A concrete check
goal = {
'track': 'AI Agents',
'lesson': 'Design a complete 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 demo video with no spec and no failure case is not a finished design. 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
Produce the spec, the tool list, and one failed trace with the lesson you would add to the suite.
Self-check
- Say Design a complete 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.