---
title: "Comment by Brian · AI News on \"The Tests Passed. The Architecture Didn't.\" | Hashnode"
description: "Observability without traces is guesswork. I log tool args latency and which policy gate fired so a bad run is a five minute debug not a day."
url: "https://hashnode.com/posts/the-tests-passed-the-architecture-didnt/6ab36ebf557667fd1ce951f0/comment/6ab43e7a2fd7b44d37cb548f"
---

[Discussion](https://hashnode.com/posts/the-tests-passed-the-architecture-didnt/6ab36ebf557667fd1ce951f0)

Comment [The Tests Passed. The Architecture Didn't.](https://hashnode.com/posts/the-tests-passed-the-architecture-didnt/6ab36ebf557667fd1ce951f0)

[Brian · AI News](https://hashnode.com/@brianainews)

[Sep 23](https://hashnode.com/posts/the-tests-passed-the-architecture-didnt/6ab36ebf557667fd1ce951f0/comment/6ab43e7a2fd7b44d37cb548f)

Observability without traces is guesswork. I log tool args latency and which policy gate fired so a bad run is a five minute debug not a day.

[Vlad Zoff](https://hashnode.com/@vladzoff)

[Sep 24](https://hashnode.com/posts/the-tests-passed-the-architecture-didnt/6ab36ebf557667fd1ce951f0/comment/6ab43e7a2fd7b44d37cb548f#reply-6ab4a5a6e85d6794c64671eb)

Yeah, logging which policy gate fired is really useful. Otherwise you just know that the run failed and end up digging through the whole thing to figure out why. Tool args and latency are useful for the same reason.