Explore Example Project

This tutorial ships as a single Bruin project with two pipelines side by side:

• chess-basic/ — a minimal "hello world" pipeline: connections, one pipeline.yml, a couple of ingestr assets, and a single SQL report. A great first look at Bruin.
• chess-advance/ — the same chess dataset expanded into a tour of Bruin's features: sensors, seeds, Python materialization, incremental SQL, the Python SDK, column & custom checks, custom variables, scheduling, notifications, and dashboards.

Both pipelines share the root .bruin.yml (connections) and .gitignore. Browse the tree on the left to explore each file.

chess-project/

.bruin.yml

.gitignore

chess-advance

chess-basic

pipeline.yml

assets

raw

games.asset.yml

profiles.asset.yml

reports

player_summary.sql

.bruin.yml

default_environment: default
environments:
  default:
    connections:
      duckdb:
        - name: "duckdb-default"
          path: "duckdb.db"
      chess:
        - name: "chess-default"
          players:
            - "FabianoCaruana"
            - "Hikaru"
            - "MagnusCarlsen"
            - "GothamChess"
            - "DanielNaroditsky"
            - "AnishGiri"
            - "Firouzja2003"
            - "LevonAronian"
            - "WesleySo"
            - "GarryKasparov"

What's inside chess-basic/

• pipeline.yml — name: chess_basic, with type: ingestr as the default asset type.
• assets/raw/games.asset.yml, profiles.asset.yml — Ingestr assets that pull the Chess.com games and profiles endpoints into DuckDB.
• assets/reports/player_summary.sql — A single SQL report that joins and aggregates into a player-level win-rate table with one column check.

That's it — enough to run an end-to-end ingest + transform flow locally.

What's inside chess-advance/

• pipeline.yml — name: chess_advance, plus pipeline-level schedule, retries, concurrency, notifications, tags, default_connections, and custom variables (min_games, rating_category).
• assets/raw/ — Raw landing layer: ingestion and seeds.
  • ingestr_games.asset.yml, ingestr_profiles.asset.yml — Ingestr assets that inherit type: ingestr from pipeline defaults.
  • seed_top_players.asset.yml (+ seed_top_players.csv) — duckdb.seed loading static CSV data, with an accepted_values column check.
• assets/sensor/ — Readiness gates.
  • sensor_games_ready.asset.yml — duckdb.sensor.query that polls until games have landed before downstream work runs.
• assets/transformations/ — Intermediate processing.
  • python_materialization_player_ratings.py — Python asset with materialize() that returns a DataFrame to be written as a table; also reads the rating_category custom variable from BRUIN_VARS.
  • incremental_sql_daily_games.sql — Incremental SQL (delete+insert keyed on game_date) showcasing the built-in {{ start_date }} / {{ end_date }} variables.
• assets/reports/ — Consumer-facing outputs.
  • sql_with_checks_player_summary.sql — Aggregate table demonstrating column checks (not_null, unique, positive, min, max), custom_checks with value / blocking, and the {{ var.min_games }} custom variable.
  • python_sdk_rating_insights.py — Bruin Python SDK + Python materialization combined. Uses from bruin import query, context to pull data from the warehouse and returns the enriched DataFrame from materialize() so Bruin writes it back as a table. Reads both built-in (context.start_date) and custom (context.vars["min_games"]) variables.
  • dashboard_chess_overview.asset.yml — tableau dashboard asset that appears as a terminal node in lineage.

Setup

Fill in .bruin.yml with your connections. You can read more about connections here.

Both pipelines share the pre-configured list of 10 popular Chess.com players. You can modify the players list in your chess connection to track different players.

Running a Pipeline

Run a whole pipeline by pointing at its folder:

bruin run chess-basic
# or
bruin run chess-advance

Or run a single asset:

bruin run chess-advance/assets/reports/sql_with_checks_player_summary.sql

Override a custom variable for a single run (chess-advance only):

bruin run --var min_games=250 --var rating_category=blitz chess-advance

You can optionally pass a --downstream flag to run an asset with all of its downstreams.