devops-ai-guidelines: An 18-Month Learning Path from DevOps to AI Infrastructure Architect

What devops-ai-guidelines Covers and Who It Is For

devops-ai-guidelines is a documentation-first repository. It contains markdown guides, code walkthroughs, and structured learning paths, not a deployable application. The stated audience covers three tiers: individual DevOps engineers who want daily productivity gains from AI tools, team leads who want a framework for safe team-wide AI adoption, and organisations or CTOs who need enterprise rollout guidelines.

The README describes its scope as an 18-month journey from DevOps Engineer to AI Infrastructure Architect. That framing is aspirational: the repository organises its material by topic across numbered folders, and a learner can work through each at their own pace. The route is not locked to 18 months, and a practitioner already comfortable with Go or Kubernetes can skip the foundational sections and go directly to the MCP or agent modules.

The repository is sponsored by Versus Incident, the maintainer's own incident management product. That context explains why the monitoring and SRE modules receive particular depth compared to the general AI roadmap sections.

Repository Layout and Available Modules

The top-level directory holds six numbered content folders, a resources folder, and standard repository files. As of the README, all six content modules are marked as available:

- 01-ai-roadmap-for-devops: the complete visual learning path with three phases and navigation aids. - 02-mcp-for-devops: a guide to building MCP (Model Context Protocol) servers using Go and Kubernetes. - 03-ai-agent-for-devops: a guide to building AI agents using Go and LangChain. - 04-ai-agent-for-monitoring: the Versus SRE Agent, which monitors systems and escalates unexpected issues by learning a baseline of normal behaviour. - 05-ai-project-management: a walkthrough of building an intelligent project management system using the OpenClaw project. - 06-sre-agent-brain: a system for storing runbook knowledge so an SRE agent can retrieve the right fix when an incident fires.

The resources folder holds standalone documents: a team AI guidelines framework, a mock interview guide, a prompt collection for daily DevOps tasks, and an AWS certification guide. Read times listed in the README range from five minutes for the prompt list to twenty minutes for the team guidelines.

How to Navigate and Use the Repository

The repository is cloned from GitHub and read locally or in a web browser. There is no install step and no application to run. Each numbered folder contains a table-of-contents file (typically 00-toc.md or 00-contents.md) that links to the individual guide pages within that module.

The README gives three distinct starting points based on role. An individual engineer is directed to start with the ten-prompt resource for immediate gains, then the AWS learning guide, and then the interview prep guide. A team lead is directed to start with the team guidelines document. An organisation deploying AI across multiple teams is directed to implement the guidelines first, then use all resources as training material.

The two example documents in the resources folder show a conversational walkthrough of using an AI assistant to build AWS infrastructure from natural-language prompts. These are illustrative, not runnable code.

What the Repository Does Not Include

devops-ai-guidelines does not include a working SRE agent, a deployable MCP server binary, or any runnable code that a reader can clone and execute without first building or extending it. The guides describe how to build these systems with referenced tools like Go, LangChain, and Kubernetes, but the code shown is instructional.

The repository has no GitHub releases, which means there is no versioned artifact to pin. The learning material evolves as the maintainers add or update modules, and there is no formal changelog for readers to track what changed between visits.

The content is primarily written for engineers comfortable reading English-language technical documentation. The repository README is in English, though one module reference (05-ai-project-management) mentions OpenClaw, an external project with its own documentation.

The MCP and AI Agent Focus

Two modules stand out for engineers doing hands-on infrastructure work. Module 02 covers MCP (Model Context Protocol), which is the open standard for connecting AI models to external tools and data sources. The guide targets DevOps engineers who want to build custom MCP servers in Go that expose Kubernetes operations, deployment pipelines, or monitoring APIs to an AI model.

Module 03 covers AI agents built with Go and LangChain. LangChain is a Python and JavaScript framework for composing LLM-based workflows. The guide walks through constructing an agent that can take actions in a DevOps context rather than just answering questions.

Module 04, the Versus SRE Agent, extends that to incident response: the agent learns what normal system behaviour looks like and escalates when it detects something outside that baseline. The module is described as complete and available in the README.

Comparison with Platform-Specific AI Tools

devops-ai-guidelines is a learning resource, not a product. A direct comparison against a tool like Datadog or PagerDuty would not be meaningful. The closer comparison is to other community-maintained AI DevOps curricula, such as the fast.ai practical courses, which focus on machine learning model training rather than infrastructure automation.

The distinction is scope. fast.ai teaches how to train and fine-tune models. devops-ai-guidelines teaches how to wire existing AI services into DevOps workflows without requiring model training. An engineer who needs to understand the ML side of AI infrastructure would need additional resources beyond what this repository covers.

Maintenance Status, Sponsorship, and Licence

The last push to the repository was on 2026-09-24, and the repository is not archived. The project has no GitHub releases. The README notes that the repository is sponsored by Versus Incident, a project from the same organisation, and includes a GitHub Sponsors link for continued development.

The project is licensed under the MIT License, which permits use, modification, and redistribution with attribution. There are no restrictions on commercial use of the materials.

Editorial conclusion

devops-ai-guidelines suits DevOps engineers who want a structured, multi-month path to building AI-assisted infrastructure tools, particularly those interested in MCP server development in Go, LangChain-based AI agents, and SRE agent design. It is not suited to anyone looking for a self-contained executable tool: there is nothing to install or run. The value is in the written guides and code examples inside each numbered folder. Before using it, check whether the section you need is marked Available in the README table, since the roadmap lists future modules that have not been published yet.

Frequently asked questions

Official sources

1. Issues
2. License: MIT
3. Project website
4. README
5. VersusControl/devops-ai-guidelines on GitHub