Agent Engineer Course

An open, practical learning path for enterprise Agent engineering, Coding Agents, model tooling, deployment, observability, evaluation, and security.

Mobile / tablet: open the course site first. It provides navigation, full-text search, previous/next lesson links, and a better reading layout than the GitHub file view.

Start learning

• Open the online course
• Course roadmap
• Continue with the latest published lesson: Lesson 12 — Kubernetes RBAC
• Source-code study — Codex, OpenCode, Kimi Code, and Claude Code public mechanisms
• Model engineering — Hugging Face and model-engineering topics
• Labs

Current published progress

Lessons 1–12 are currently in the repository. Lesson 12 introduces Kubernetes ServiceAccount, Role/RoleBinding, Secret boundaries, SecurityContext, and least privilege for Agent Scheduler and Runner workloads. Next is NetworkPolicy and Agent egress control.

Local preview

python -m pip install mkdocs-material
mkdocs serve

Then open the local address printed by MkDocs in a desktop browser.

Public-content policy

Examples use generic names such as sample-app, demo-repo, and placeholder paths. Do not add personal projects, credentials, private paths, account-specific data, or secrets to course examples.

Publishing

A GitHub Actions workflow in .github/workflows/deploy.yml builds the MkDocs Material site from main and deploys it to GitHub Pages. If Pages is not yet visible, repository Settings → Pages may need GitHub Actions selected as the source.

License and disclaimer

Educational material only. Verify current product behavior and official documentation before using these ideas in production. See Disclaimer.