awsiamPractitioner

Automating Incident Management with AWS DevOps Agent: A Case Study

5 min read AWS DevOps BlogSep 28, 2026Reviewed for accuracy

Practitioner — Hands-on experience recommended

In the fast-paced world of software development, managing incidents efficiently is crucial. Property Finder faced challenges in handling incidents manually, which often led to delays and inconsistencies. By leveraging the AWS DevOps Agent, they automated their incident management workflows, allowing for quicker resolutions and less human error.

The process kicks off when an 'Investigation Completed' event is fired by Amazon EventBridge. This triggers a Lambda orchestrator that invokes the remediation agent using the investigation ID. The agent dives into journal records to uncover the root cause and recommended fixes. It then maps the AWS account to the appropriate repository, which is essential given Property Finder's six different infrastructure repositories for various teams. The agent checks for existing pull requests (PRs) to avoid duplicates, reads relevant Terraform files, and identifies necessary changes. Finally, it creates a feature branch, commits the changes, and opens a Draft PR complete with a structured template that includes a problem summary, root cause, changes made, and a testing checklist.

In production, the key to success lies in the agent's ability to validate repository mappings through automated tests and adapt as new accounts are onboarded. This ensures that the incident management process remains robust and scalable. However, keep in mind that while automation can significantly enhance efficiency, it requires careful monitoring and maintenance to avoid potential pitfalls.

Key takeaways

→Automate incident workflows using the AWS DevOps Agent to reduce manual errors.
→Utilize Amazon EventBridge to trigger incident management processes seamlessly.
→Implement checks for duplicate PRs to maintain a clean codebase.
→Structure Draft PRs with templates to ensure clarity and thoroughness in communication.
→Adapt repository mappings through automated tests to keep pace with changes.

Why it matters

Automating incident management with the AWS DevOps Agent can drastically reduce response times and improve the reliability of incident resolutions, leading to enhanced overall system stability.

Code examples

Python

1# HMAC-SHA256 signing for webhook authentication
2ts = datetime.now(timezone.utc).strftime("%Y-%m-%dT%H:%M:%S.000Z")
3body = json.dumps(incident)
4sig = hmac.new(webhook_secret.encode("utf-8"),
5               f"{ts}:{body}".encode(), hashlib.sha256).digest()
6http.request("POST", webhook_url, body=body,
7             headers={"x-amzn-event-timestamp": ts,
8                      "x-amzn-event-signature": base64.b64encode(sig).decode()})