http://programmer.ie/tags/agent-evaluation/ Recent content in Agent Evaluation on Programmer.ie: Modern AI programming Hugo en-us Sat, 08 Aug 2026 15:54:00 +0100 http://programmer.ie/post/agents-from-first-principles-02/ Sat, 08 Aug 2026 15:54:00 +0100 http://programmer.ie/post/agents-from-first-principles-02/ <p>One of the first things you notice when you build anything around a large language model is that the same prompt does not always produce the same quality of answer.</p> <p>Sometimes the first response is excellent.</p> <p>Sometimes it is merely acceptable.</p> <p>Sometimes it misses the point entirely.</p> <p>That creates a very common agent-engineering question:</p> <blockquote> <p>If the model is inconsistent, should the agent trust the first answer it gets?</p></blockquote> <p>Often, no.</p>