C.W.K.
Stream
← C.W.K. Quests
🏗️

CI CD Quest

Updated: 2026-05-04

Automate everything with GitHub Actions

From the moment you push a commit to the moment the world sees it — every gate, every test, every deploy is a workflow you wrote. CI CD Quest teaches you to build that pipeline with GitHub Actions, the way real engineering teams ship in 2026.

8 tracks · 101 lessons · ~50h · difficulty: beginner-to-advanced

Level 0Apprentice
0 XP0/101 lessons0/10 achievements
0/120 XP to next level120 XP to go0% complete
CI/CD is not a tool category. It is the discipline of letting code prove itself before it ships. This quest walks the full arc — what continuous integration and delivery actually mean, how GitHub Actions models the world (workflows, jobs, steps, runners), how to build real test and deploy pipelines for Python, Node, Docker, AWS, Vercel; how secrets, OIDC, and environments guard production; how reusable workflows and composite actions keep the YAML from rotting; how AI/ML projects use CI for prompt regression and eval gates; and how self-hosted runners turn your own hardware into a fleet. Eight tracks, 101 lessons, every example pulled from real repos that ship today — including the very repo you are reading this in.

Tracks

  1. 01🏗️CI/CD Foundations

    0/12 lessons

    What integration, delivery, and deployment actually mean

    CI/CD is a discipline before it is a YAML file. This track grounds the vocabulary, explains why the pre-CI/CD world hurt, and shows where GitHub Actions sits in the broader landscape. By the end you can defend each acronym and know when not to use any of it.

    Lesson list (12)Quiz · 5 questions
  2. 02🔧GitHub Actions Core Concepts

    0/12 lessons

    Workflows, jobs, steps, runners, actions

    GitHub Actions has six primitives that show up in every workflow you'll ever write. Once you can read them like sentences, the rest of the system reduces to composition. This track makes those primitives second nature.

    Lesson list (12)Quiz · 5 questions
  3. 03🛡️Building & Testing Pipelines

    0/13 lessons

    Lint, type-check, test, scan — the gates that keep main green

    This is the meat of CI. Real test pipelines for Python and Node, with caching that takes minutes off every run, matrices that catch regressions across Python/Node versions, lint and type-check gates that fail before the test even starts, and security scanners that find secrets in your code. By the end you can wire any project into a real test pipeline.

    Lesson list (13)Quiz · 5 questions
  4. 04🔐Secrets, Environments & Variables

    0/12 lessons

    How to ship without long-lived keys leaking everywhere

    Workflows that touch real systems need credentials. This track is about how to give them just enough — repository / org / environment-scoped secrets, environment protection rules, GITHUB_TOKEN scopes, and OIDC for keyless cloud auth. By the end your pipelines deploy without any long-lived static credentials.

    Lesson list (12)Quiz · 5 questions
  5. 05♻️Advanced Workflows & Reuse

    0/14 lessons

    Job graphs, reusable workflows, composite actions, dynamic matrices

    Once your workflow file passes 200 lines, it stops being readable. This track shows the seven patterns that keep CI YAML small even as the system grows: <code>needs:</code> graphs, <code>if:</code> conditions, reusable workflows, composite actions, concurrency, dynamic matrices, and triggers like <code>workflow_dispatch</code>, <code>schedule</code>, and <code>repository_dispatch</code>.

    Lesson list (14)Quiz · 5 questions
  6. 06🚀Deployment Pipelines

    0/13 lessons

    From a green build to live in production

    This track is the actual ship. Strategies (rolling, blue/green, canary), real targets (Pages, Docker registries, AWS, GCP, Azure, Vercel, SSH), the release machinery (tags, version bumps, changelogs), the disasters (database migrations, rollback), and the smoke tests that turn 'deploy succeeded' into 'production actually works'.

    Lesson list (13)Quiz · 5 questions
  7. 07🤖CI/CD for AI/ML Projects

    0/12 lessons

    Eval gates, prompt regressions, GPU runners, model registries

    AI/ML changes are tricky in CI: behavior is statistical, model files are huge, GPUs are expensive, and 'a regression' isn't always a thrown exception. This track covers the patterns Pippa actually uses — LLM eval gates, prompt regression suites, notebook testing, large-file handling, GPU runners, and the eval-gate pattern that keeps prompt changes from quietly degrading production.

    Lesson list (12)Quiz · 5 questions
  8. 08⚙️Self-Hosted Runners & Advanced Ops

    0/13 lessons

    Run your own CI infrastructure at scale and observe what it does

    When hosted runners stop being enough — too expensive, too small, missing GPUs or Apple Silicon, or you need access to a private network — you bring your own runners. This track covers the full self-hosted lifecycle: setup, security, debugging, performance tuning, monitoring, GitHub Apps for richer integration, and the alternative platforms worth comparing. The final lesson is the ideal CI setup as Pippa would build it from scratch.

    Lesson list (13)Quiz · 5 questions
Spotted a bug or have feedback on this page?Report an Issue

Comments 0

🔔 Reply notifications (sign in)
Sign inPlease sign in to comment.

No comments yet — be the first.