DataTalksClub/data-engineering-zoomcamp ? reverse-engineered prompt

I want to build a project to learn data engineering from scratch. I'm following the syllabus from the popular "Data Engineering Zoomcamp" course on GitHub.

The goal is to create a complete, end-to-end data pipeline. It should be set up to use Docker and Terraform for the infrastructure, maybe on GCP. I'll need a workflow orchestrator like Kestra to manage the tasks. The pipeline should ingest data, move it into a data warehouse like BigQuery, and then use a tool like dbt for the analytics and transformation parts.

It looks like the course also covers batch processing with Spark and streaming with Kafka, so I'd want to include modules for those as well. Can you set up the project structure with folders and starter code for each of these stages? Please add comments to explain the key parts.