Build Your First Data Product

This guide builds hello-orders, a minimal data product with one seed, one staging model, one mart, and dbt tests. Customer 360 is the advanced demo after you understand this path.

Prerequisites

A Platform Environment Contract from your Platform Engineer.
A Floe platform installed and validated by a Platform Engineer.
A repository checkout with the examples/hello-orders project.
uv, Python, and the repository development dependencies installed for the alpha docs path.

1. Inspect The Environment Contract

sed -n '1,220p' examples/platform-environment-contracts/dev.yaml

2. Inspect The Data Product

find examples/hello-orders -maxdepth 3 -type f | sort
sed -n '1,180p' examples/hello-orders/floe.yaml

3. Review The dbt Models

sed -n '1,120p' examples/hello-orders/models/staging/stg_orders.sql
sed -n '1,120p' examples/hello-orders/models/marts/mart_daily_orders.sql
sed -n '1,180p' examples/hello-orders/models/schema.yml

4. Compile The Product For The Alpha Runtime Contract

uv run floe platform compile \
  --spec examples/hello-orders/floe.yaml \
  --manifest demo/manifest.yaml \
  --output target/hello-orders/compiled_artifacts.json \
  --generate-definitions

The root floe compile, floe run, and floe product deploy commands are planned product lifecycle entry points. They are not the current alpha path.

5. Package A Runtime Artifact

For the alpha path, CI should build a product runtime image that contains:

dbt project files.
compiled_artifacts.json.
generated Dagster definitions.
runtime dependencies pinned by the repository lockfile or organization base image.

6. Deploy Through Your Organization’s Approved Path

Use the handoff pattern documented in Data Product Runtime Artifacts. Floe does not mandate GitHub, GitLab, Jenkins, Argo CD, Flux, Backstage, or a specific registry.

7. Validate The Product

Continue with Validate Your Data Product.

8. Then Run Customer 360

After hello-orders, use Customer 360 to prove the full business demo and release-validation path.