AI for DevOps: Claude Code on Your Infrastructure, Safely
AI agents are great at the parts of ops that eat your day: reading logs, explaining a failing pipeline, drafting Terraform, reviewing a plan for the one line that replaces your database. They're dangerous at the parts that can't be undone. This course draws that line in configuration: read-only identities, deny and ask rules for destructive verbs, plan review as structured output, a triage skill for incidents and CI agents with minimal tokens. Written by a DevOps and cloud architect who works with AWS, Azure and Kubernetes.
New
Offered by ITwithDom- Last updated November 2026
- Language: English
- 5 h
- E-learning
What you'll learn
- Decide which ops tasks to hand to an agent, and which must stay behind a pipeline and a human
- Give an infra agent read-only, short-lived access, and block destructive commands with deny rules and hooks
- Review Terraform plans with Claude: structured risk output for destroys, replacements, IAM and public exposure
- Triage Kubernetes incidents with a read-only runbook skill
- Run agents in CI with least-privilege tokens, budgets and no secrets in prompts
Practical information
- Format
- E-learning
- Location
- Online
- Schedule
- Self-paced: about 1 h 30 of reading plus 3–4 h of hands-on work in a disposable environment.
Course content
7 lessons, 1 h 30 in total
- Where AI helps in ops, and where it hurts11 min
- Read-only first: identity and permissions15 min
- Terraform plan review with Claude15 min
- Kubernetes and incident triage14 min
- Agents in CI12 min
- Guardrails for infra agents11 min
- Final project: your ops copilot12 min
Description
A practical course for DevOps, platform and cloud engineers on using Claude Code with infrastructure: read-only access first, Terraform plan reviews, Kubernetes and incident triage, CI agents with least-privilege tokens, and guardrails that block destructive commands.
Key information
- Learning objectives
- Configure Claude Code for an infrastructure repository with read-only access and blocked destructive verbs
- Build a Terraform plan-review step with structured JSON output
- Write an incident-triage skill for Kubernetes
- Set up a CI agent job with minimal permissions and a budget
- Who it is for
- DevOps, SRE, platform and cloud engineers working with Terraform, Kubernetes and CI/CD. The hands-on steps use AWS and GitHub Actions; Azure, GCP and GitLab users adapt the cloud-identity and CI steps.
- Prerequisites
- Working knowledge of Terraform and kubectl, and a GitHub account (the CI exercise uses GitHub Actions). For the hands-on parts: a disposable Kubernetes cluster (kind or minikube is enough), a sandbox AWS account (exercise 2 also works with the kit's sample plan only), Claude Code 2.1.295+, jq; macOS, Linux or Windows with WSL2 (on Linux and WSL2, bubblewrap and socat for the sandbox). An Anthropic API key (Console, pay-as-you-go) is required for the CI exercise and the final project's CI job, and optional for exercise 2; each review run is capped at $1 by --max-budget-usd. AWS CLI v2 for the cloud steps.
- Duration
- 5 h
- Price
- Free
- Teaching format
- E-learning
- Access conditions and lead time
- Free. Create a learner account on the course page and start immediately. Immediate access after enrolment.
- Teaching methods
- Short explanations with diagrams and an animated walkthrough, hands-on configuration in an infrastructure repository, and a final « ops copilot » project.
- Assessment
- Short checkpoint quiz on every page (one mistake allowed), final exam at 70 %, Four hands-on exercises with a rubric, Final project with a reference solution
- Accessibility for people with disabilities
- Self-paced, no time limit. Diagrams have text descriptions, the animation has captions, all code is text.
- Contact
- Dominik Heinrich (ITwithDom)
The training provider
ITwithDomSee the profile and all its courses More courses from ITwithDom
See all
AI for DevOps: Claude Code on Your Infrastructure, Safely
Free