---
title: "Discussion on \"How to trace and evaluate an AI agent in Microsoft Foundry, from the first trace to production monitoring\" | Hashnode"
description: "Discussion on \"How to trace and evaluate an AI agent in Microsoft Foundry, from the first trace to production monitoring\". Microsoft Foundry splits agent observability into three capabilities, and each one answers a different question. Tracing shows what the agent did during a run, evaluation scores how good its answers w"
url: "https://hashnode.com/posts/how-to-trace-and-evaluate-an-ai-agent-in-microsoft-foundry-from-the-first-trace-to-production-monitoring/6ae016289da8848c6350c83d"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [How to trace and evaluate an AI agent in Microsoft Foundry, from the first trace to production monitoring](https://carlosjcastrog.hashnode.dev/how-to-trace-and-evaluate-an-ai-agent-in-microsoft-foundry-from-the-first-trace-to-production-monitoring?utm_source=hashnode&utm_medium=feed)

Microsoft Foundry splits agent observability into three capabilities, and each one answers a different question. Tracing shows what the agent did during a run, evaluation scores how good its answers w

Oct 27 · 9 min read · carlosjcastrog.hashnode.dev  
Read the article

-   [#azure](https://hashnode.com/tag/azure)
-   [#ai](https://hashnode.com/tag/ai)
-   [#cloud](https://hashnode.com/tag/cloud)
-   [#micro](https://hashnode.com/tag/micro)

## Written by

[Carlos José Castro Galante](https://hashnode.com/@carlosjcastrog)

@carlosjcastrog · Argentina

  
[View profile](https://hashnode.com/@carlosjcastrog)

Software Developer

Software Developer based in San Juan, Argentina. Microsoft Student Ambassador. AWS Student Builder Group Leader. Certified in AI-102, AI-900 and AZ-900. Writing about Azure, AI and cloud development

## Responses

Markdown supported · ⌘↵ to post

No responses yet.
