Learn to Cloud App

A web application for tracking your progress through the Learn to Cloud guide.

Note: This project is open source under the MIT License.

Features

• 📚 All 8 phases of the Learn to Cloud curriculum
• ✅ Progress tracking with steps, questions, and hands-on projects
• 🔐 Authentication via GitHub OAuth
• 📊 Dashboard with progress visualization
• 🐙 GitHub integration for project submissions
• ⚙️ Background verification inside the API container

Tech Stack

GitHub login establishes revocable, PostgreSQL-backed sessions. See Authentication and sessions for route dependencies, login redirects, expiry, and logout guarantees.

Quick Start

WSL / Linux Setup

Development runs directly in WSL or Linux. On Windows, install WSL 2, keep the clone inside the WSL filesystem, and make Docker Desktop's WSL integration available to that distribution.

Prerequisites

• Git
• uv
• Docker with the Compose plugin
• Node.js 20+ for the full test suite (not required just to start the API)

uv installs the required Python 3.13 runtime. Frontend, verification, infrastructure, and agent workflows need additional optional tools documented in the Contributing Guide.

Local Development

1. Start local dependencies (Docker)

docker compose up -d db aspire-dashboard

2. Install Python dependencies

Install the app and its development tools into a virtual environment:

uv sync --locked
cp .env.example .env  # Create environment config (edit if needed)

Run database migrations:

uv run alembic upgrade head

Start the API:

uv run python -m uvicorn learn_to_cloud.main:app --reload --port 8000

Or use VS Code's debugger with the "API: FastAPI (uvicorn)" launch configuration.

The API starts verification automatically: one sequential background loop per API process atomically claims attempts from PostgreSQL. No separate host, external queue, or verification job is required. Attempts have an execution timeout and overdue cleanup, with no workflow retries or checkpoints.

Notes:

• The API does not start local dependencies for you. Run docker compose up -d db aspire-dashboard first.
• Manage dependencies with docker compose start / docker compose stop.

Project Structure

├── src/
│   └── learn_to_cloud/   # FastAPI app (serves HTML + JSON API)
│       ├── main.py       # App entry point
│       ├── routes/       # API + page endpoints
│       ├── services/     # Business logic
│       ├── core/         # Config, auth, database, telemetry
│       ├── repositories/ # Database access
│       ├── verification/ # Verification checks and engine
│       ├── content/      # Curriculum YAML and compiled JSON
│       ├── migrations/   # Alembic migrations (shipped in the package)
│       ├── templates/    # Jinja2 templates (HTMX)
│       └── static/       # CSS, JS, images
├── tests/                # pytest suite
├── scripts/              # Dev and operator scripts
├── docs/                 # Contributor docs
├── infra/                # Terraform (Azure)
├── Dockerfile            # API and migration images
├── pyproject.toml        # Dependencies, tool config, poe tasks
└── .github/
    ├── workflows/        # CI/CD
    ├── copilot-instructions.md # Copilot custom instructions
    └── skills/           # Copilot agent skills

Contributing

Start with the Contributing Guide for setup, quality gates, and links to focused development guides.

Deployment

CI runs on pull requests. Pushes to main that change infrastructure, application runtime files, or the deploy workflow run deploy.yml, one ordered pipeline that runs a single deploy at a time:

1. Plans Terraform and applies it when there are changes.
2. Releases the API when application files (src/, Dockerfile, dependency manifests) differ from the commit production currently runs. It builds and pushes the API and migration images, runs migrations, and updates the API.
3. Verifies that the latest revision runs the expected image, is healthy, and returns 200 from /ready.

Each run releases the latest main, not just the commit that triggered it, so a run that waited behind another deploy cannot roll production back. Tests and documentation do not deploy.

Ship infrastructure and application changes as separate pull requests, infrastructure first, so either can be reverted on its own.

A manual Deploy run redeploys the current main; select force_rebuild to rebuild the API without cache. Select plan_only to plan Terraform for the selected branch without applying or deploying.

After a failed deployment, fix and rerun it before shipping another release. Do not use an old workflow run as an image-only rollback.

License

MIT License. See LICENSE.