# Create Expert Content: Deploying a Multi-Agent System with Terraform and Cloud Run

Canonical URL: https://www.traeai.com/articles/1db9a34e-915a-4ef6-9716-4f6a7f5ea814
Original source: https://cloud.google.com/blog/topics/developers-practitioners/create-expert-content-deploying-a-multi-agent-system-with-terraform-and-cloud-run/
Source name: Google Cloud Blog
Content type: article
Language: 英文
Score: 7.5
Reading time: 8 分钟
Published: 2026-04-17T08:56:00+00:00
Tags: Google Cloud, Cloud Run, Terraform, 多智能体系统, MLOps

## Summary

文章详解如何用 Terraform 和 Cloud Run 将多智能体系统 Dev Signal 从本地原型部署到生产环境，强调可观测性、内存持久化与安全基础设施。

## Key Takeaways

- 使用 FastAPI 构建应用服务器作为多智能体系统的生产入口，连接 Vertex AI 内存实现状态持久化。
- 通过 ADK 的 otel_to_cloud=True 自动导出 Agent Traces 到 Cloud Console，实现智能体推理过程可视化。
- Terraform 自动化创建 Artifact Registry、最小权限服务账号和 Secret Manager，保障部署安全性与可复现性。

## Citation Guidance

When citing this item, prefer the canonical traeai article URL for the AI-readable summary and include the original source URL when discussing the underlying source material.