Analytics¶

Orchestrate pipelines with Prefect or Airflow, transform data with dbt, and visualize results with Grafana or Metabase.

Best for: scheduled workflows with monitoring, BI dashboards, data observability.

Tools¶

Phase	Options
Orchestrate	Prefect · Airflow
Transform	dbt
Store	Postgres · ClickHouse
Serve	Grafana · Metabase

Create the project¶

degen init

Select Analytics. This guide uses Prefect + dbt + Postgres + Grafana.

cd my_analytics_project

Install¶

degen install

Installs: prefect, dbt-core, dbt-postgres, psycopg2-binary, and scaffolds the dbt project.

Start services¶

degen docker-up

Starts three containers:

Container	Service	Port
`degen_postgres`	PostgreSQL 16	5432
`degen_pgadmin`	pgAdmin 4	5050
`degen_grafana`	Grafana	3001

Wait ~10 seconds for all services to become healthy:

degen status

Docker services:
NAME              STATUS
degen_postgres    Up (healthy)
degen_pgadmin     Up
degen_grafana     Up (healthy)

Start Prefect server¶

Open a dedicated terminal and run:

degen prefect-server

This starts the Prefect backend. Navigate to http://localhost:4200 to open the Prefect UI.

Run the pipeline¶

In another terminal:

degen run

The Prefect flow runs three tasks in sequence:

INFO  Flow run 'bold-falcon' - Beginning flow run for flow 'degen-pipeline'
INFO  Task run 'extract-812' - Extracting data... Extracted 5 records
INFO  Task run 'transform-42c' - Transforming data... Transformed 5 records
INFO  Task run 'load-fd1' - Loading data... Load complete.
INFO  Flow run 'bold-falcon' - Finished in state Completed()

Validate in Prefect UI¶

Open http://localhost:4200

Flow Runs → see all executions with timing and status
Flows → registered flows
Tasks → individual task runs with logs

Each run shows the full execution tree: extract → transform → load

pgAdmin — browse Postgres¶

Open http://localhost:5050

Field	Value
Login	`admin@degen.io`
Password	`degen`

The server "DEGEN Postgres" is pre-configured. Click it and enter the database password:

Field	Value
Password	`degen`

Navigate to: Servers → DEGEN Postgres → Databases → degen → Schemas → public → Tables

Grafana — build dashboards¶

Open http://localhost:3001

Field	Value
Login	`admin`
Password	`admin`

Add Postgres as a data source:

Connections → Data Sources → Add → PostgreSQL
Fill in:

Field	Value
Host	`postgres:5432`
Database	`degen`
User	`degen`
Password	`degen`
TLS/SSL	Disable

Click Save & Test
Create dashboards with SQL queries against your dbt models

dbt documentation¶

degen docs

Serves the dbt lineage graph and model docs at http://localhost:8081.

Using Airflow instead of Prefect¶

Select Airflow during degen init. DEGEN configures 4 containers:

Container	Role
`airflow-postgres`	Airflow metadata DB
`airflow-init`	DB migration + admin user
`airflow-webserver`	UI at :8080 (admin/admin)
`airflow-scheduler`	DAG scheduler

degen docker-up       # starts all 4 Airflow services
degen orchestrate     # prints Airflow UI URL

A sample DAG with extract → transform tasks is pre-generated in dags/pipeline_dag.py.

Project structure¶

my_analytics_project/
├── degen.yaml
├── .env
├── profiles.yml            # Postgres connection with env_var()
├── docker-compose.yml      # Postgres + pgAdmin + Grafana
├── requirements.txt
├── flows/
│   └── pipeline_flow.py    # Prefect flow (or dags/ for Airflow)
└── my_analytics_project/   # dbt project
    ├── dbt_project.yml
    └── models/