Season 8: You Own the Release · Episode 2 of 8
From Commit to Candidate
You play: AI Engineering Intern
Included with Core programme and Certification prep
The situation
assist-service used to ship by someone remembering the steps: install, run the tests, build it, hope. Kabelo has had enough of hoping. Today the repo gets a real pipeline — install, test, build, containerize — and every commit that reaches main runs through the same stages, the same way, every time.
What you'll learn
- CI/CD stages and explicit dependencies; triggers; dependency-cache keys; artifact and commit traceability; inspect repeat-build evidence while accounting for mutable external inputs and environment differences.
- GitHub Actions versus Azure Pipelines vocabulary (on/trigger, needs/dependsOn, runs-on/pool); reading a workflow for job order; Azure Artifacts and GitHub Packages, feeds, views and upstream sources; SemVer and CalVer; code coverage and load tests; hosted versus self-hosted runners and agents; moving classic pipelines to YAML; flaky tests, duration and retention.
Who you work with
Kabelo Mokoena
Platform and DevOps Engineer
Maya Chen
Senior AI Engineer · Your mentor
Scenes
- 1.Stop hoping, start running the same way every time
- 2.Stage, trigger, artifact, or cache?
- 3.Actions or Azure Pipelines?
- 4.Read the workflow
- 5.A commit, start to candidate
- 6.Broken, fixed, and what the cache remembers
- 7.Prove it builds the same thing twice
- 8.Green isn't the same as verified
- 9.Handover
What you leave with
You exercised the lab's install–test–build–containerize pipeline, failure gates, recorded triggers and dependency cache. Two runs produced matching fixture digests, and you selected that content reference over a movable tag. This is a simulator: real pipelines need explicit job dependencies, product-specific cache behaviour, controlled build inputs and provenance checks. Matching two digests does not establish universal reproducibility or safety.