Asked about What an agent actually is
When would you not use an agent?
An answer that works
When the steps are known in advance. A fixed pipeline is cheaper, faster and far easier to debug. Agents earn their cost only when the path depends on what is discovered along the way.
Then they push further
Give an example of a task that looks like it needs an agent but does not.
Summarising a support ticket and routing it to a team. It feels open-ended, but the steps are fixed: classify, summarise, route. Two calls with structured output do it cheaper, and every step can be tested on its own. An agent would add variance and cost for no gain.
And one that genuinely does?
Working out why a customer's invoice is wrong across billing, usage and plan history, where what to look up next depends on what the last lookup showed. The number and order of steps cannot be written down in advance.
What gets marked up
Strong answers name what agency costs: variance, latency, money and harder debugging, and treat an agent as something a task has to earn. Weak answers treat agents as the default modern architecture.
Reading an answer is not being able to give one. The section that teaches this is What an agent actually is, 3 hours of The agent loop.
Learn this properlyAlso asked about The agent loop