Automation lets systems execute repeatable work with less direct human effort. In IT, AI can help plan, write, document, and extend an automated process, shifting your role from performing every step to orchestrating the system. section 1.3
Common workplace uses include:
- Check server health, inspect logs, review configuration files, and generate scripts for tasks such as kernel-version checks or file renaming. section 9.10
- Integrate operational work with ITIL practices, including incident, problem, and change management. section 9.6
- Update large codebases and test suites, such as converting 7,000 Pester tests while controlling tool and processing costs. section 12.0 section 12.5
- Reduce communication overhead by summarizing messages, identifying action items, suggesting responses, and creating simple integrations. section 7.2
Automating HR data cleanup
An HR team manually extracted information from messy vendor CSV files and copied it into a SharePoint list. A Power App triggered Azure OpenAI Service, which used function calling to extract and clean the data before updating the list. The workflow was tested with sample data, refined with HR feedback, and documented before deployment. Validation, access controls, and query checks remain essential when automation handles business data. section 13.3.3 section 13.6
Table 9.1 Systems engineering task prompts
|
Task
|
Prompt
|
|---|---|
|
Business process integration
|
I need help integrating ITIL into our Windows systems team’s daily work. Could you help me set up ITIL-based incident, problem, and change management processes? Please focus on practical steps for a Windows environment. Ask me anything you need to know about our current setup.
|
|
Network configuration generator
|
Write a network config script to set up [a VLAN network] in vSphere. Include [VLAN IDs, IP schemes, subnetting, gateway setup, and security]. The script needs to match our network standards and security rules. I’ve attached our current network requirements - let me know if you need any details clarified.
|
|
Troubleshoot container orchestration problems
|
List the key diagnostic commands for fixing common Kubernetes problems like pod failures, service discovery issues, and resource limits. Make the steps clear enough for junior admins, focusing on using kubectl for diagnosis and fixes.
|
|
System update planner
|
Create a step-by-step plan for [updating our server farm]. Cover [update order, backups, testing, and rollback steps]. The plan should keep downtime low and follow our IT standards and risk policies.
|
|
Kerberos Troubleshooter
|
Help me build a troubleshooting guide for fixing random Kerberos auth issues, focusing on "KRB_AP_ERR_MODIFIED" errors. Include steps to [check clock sync, review service principal names, and examine TGT lifetimes]. Give clear starting points and expected results that follow Kerberos best practices.
|
Use the task-prompt table to adapt automation to business-process integration, network configuration, and container troubleshooting. section 9.6
Treat generated automation as a starting point, not a guarantee. Clarify requirements, test on representative data, inspect outputs, and preserve security and operational checks before wider deployment. section 13.3.3 section 13.6
