Capstones
Cross-discipline capstones: apply the Advanced Python engineering discipline end-to-end to a real artefact from another discipline - the proof that the skills transfer.
Every pillar in this discipline can be passed on its own material, which leaves one question open: does any of it transfer? Packaging a package built to be packaged is not evidence. The test is whether you can take something that came from somewhere else, made by someone with different priorities, and apply the whole engineering discipline to it end to end.
That is what this domain is for. The artefact comes from another discipline entirely — a trained tabular model from Data Science & AI — and it is given to you, because the point is emphatically not the modelling. The work is everything around it: make it installable and reproducible, put a CI gate in front of it with a quality floor that a regression actually trips, and make its serving observable so that a slow or wrong prediction can be investigated rather than guessed at. It runs entirely in Pyodide with the production simulators inlined, so the capstone is a real exercise rather than a description of one. Finishing it is the discipline's own evidence that these are engineering skills and not Python trivia.