INDEX

• Basic @dg.asset — simple one-to-one transformation
• Parameter-based dependency — asset depends on another managed asset, data loaded via IOManager
• deps= dependency — asset depends on external or non-Python asset, data dependency only
• @multi_asset — single operation produces multiple related assets
• @graph_asset — multiple op steps needed to produce one asset
• @graph_multi_asset — complex pipeline producing multiple assets from composed ops
• Asset factory — generate many similar assets programmatically

Quick Notes on Basic Patterns

Basic @dg.asset: Function name becomes the asset key. Docstring becomes the description in the UI. Return type annotation is optional but recommended.

Asset groups: Use group_name= on the decorator to organize assets visually in the UI. Common groupings: by data layer (raw, staging, analytics), by domain (sales, marketing), or by source (postgres, api).

Key prefixes: Use key_prefix=["warehouse", "raw"] to namespace asset keys hierarchically (e.g. warehouse/raw/orders). Useful for multi-tenant or layered architectures.

Configuration: Subclass dg.Config with typed fields, then add as a parameter to your asset function. Fields become configurable at launch time.

Execution context: Add context: dg.AssetExecutionContext as a parameter to access context.log, context.asset_key, context.partition_key (if partitioned), and context.run_id.

Return types: Assets can return data directly (passed to downstream via IOManager) or dg.MaterializeResult (for metadata, or dg.MaterializeResult[T] for data + metadata). MaterializeResult[T] is preferred over dg.Output[T] in greenfield code.

Common Anti-Patterns

• Verb-based names like load_customers — use nouns describing the output: customers
• Giant asset doing everything — split into focused, composable assets
• No type annotations — add a return type: -> dict, -> None
• No docstring — add a description via docstring or description=
• Ignoring MaterializeResult — return metadata for observability

Reference Files

• Advanced Asset Patterns — @multi_asset, @graph_asset, @graph_multi_asset, or asset factories; MaterializeResult, dynamic metadata, MetadataValue types
• Asset Definition Properties — asset metadata, tags, owners, groups, key_prefix, code_version, AssetSpec properties
• Asset Dependencies — asset dependencies, parameter-based deps, deps= external dependencies