azureml-scaffolding

name	azureml-scaffolding
description	Scaffold, structure, and manage AI/ML projects that run on AzureML. Covers project initialization (uv workspaces, devcontainers, Makefile), Python packaging with explicit dependencies, local and cloud execution, experiment reproducibility, and extensibility patterns (pipelines, datasets, linting). Use this skill whenever the user asks to create, modify, run, test, or deploy an AzureML-based ML project — or when they need guidance on project layout, dependency management, or cloud job submission with Azure Machine Learning.
metadata	{"category":"data","source":{"repository":"https://github.com/bepuca/azureml-scaffolding","path":"azureml-scaffolding","license_path":"LICENSE","commit":"f7a95c6dbe77dcc15ee545bf25e87cc5bb802cde"}}

AzureML Project Scaffolding

A battle-tested structure for AI projects that require reproducible experimentation, leveraging AzureML for cloud execution. It ensures reproducibility from day one without sacrificing the path to production — and without breaking the ability to keep experimenting once you're there. Code, environments, specs, and dependencies are wired so that what runs locally runs on AzureML, with no surprises.

Principles

These principles are foundational. Every decision about project structure, tooling, or workflow must be evaluated against them.

• Three layers — Each layer depends only on inner layers:

  1. Code — the what. Pure Python, no platform deps.
  2. Specification — the how. job YAML. Declares how code executes on a target platform. Lives next to the code it describes.
  3. Orchestration — the when. Makefile, CI. Triggers execution. Knows about specs, knows nothing about code internals.