请问HN：当多步骤AI工作流程的输出错误时，您如何进行调试？
我一直在构建多步骤的人工智能工作流，涉及多个智能体（规划、推理、工具使用等），有时会遇到最终输出不正确的情况，尽管技术上没有任何故障。没有运行时错误——只是结果错误。
主要挑战在于找出问题出在哪里。问题可能出现在早期的推理步骤、步骤之间的上下文传递，或者是一个微妙的错误在系统中传播。等我看到最终输出时，哪个步骤导致了问题并不明显。
我一直在使用Langfuse进行追踪，这有助于捕捉输入和输出，但在实际操作中，我仍然需要逐步手动检查每个步骤以诊断问题，这很快就让人感到疲惫。
我很好奇其他人是如何处理这个问题的。有没有更好的方法来构建或工具化这些工作流，以便更容易定位故障？有没有什么模式、工具或技术对你们有效？
查看原文
I’ve been building multi-step AI workflows with multiple agents (planning, reasoning, tool use, etc.), and I sometimes run into cases where the final output is incorrect even though nothing technically fails. There are no runtime errors - just wrong results.<p>The main challenge is figuring out where things went wrong. The issue could be in an early reasoning step, how context is passed between steps, or a subtle mistake that propagates through the system. By the time I see the final output, it’s not obvious which step caused the problem.<p>I’ve been using Langfuse for tracing, which helps capture inputs and outputs, but in practice I still end up manually inspecting each step one by one to diagnose issues, which gets tiring quickly.<p>I’m curious how others are approaching this. Are there better ways to structure or instrument these workflows to make failures easier to localize? Any patterns, tools, or techniques that have worked well for you?