feature-store-design
Online/offline feature serving, point-in-time correctness, Feast patterns, and feature computation design.
npx skills add majiayu000/claude-skill-registry --skill feature-store-design --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# Feature Store Design ## Architecture Decision Table | Criteria | Feast (OSS) | Tecton | Vertex Feature Store | Custom (Redis + Warehouse) | |----------|------------|--------|---------------------|---------------------------| | Setup cost | Low | High (SaaS) | Medium (GCP-only) | Medium-High | | Online serving latency | <10ms (Redis) | <5ms | <10ms | Depends on impl | | Offline store | File/BigQuery/Redshift | Spark/Snowflake | BigQuery | Your warehouse | | Streaming features | Limited (push-based) | Native Spark/Flink | Dataflow | Build your own | | Point-in-time joins | Built-in | Built-in | Built-in | Must implement | | Team size sweet spot | 2-15 | 15-100+ | Any (GCP shops) | 5-20 (eng-heavy) | **Recommendation**: Feast for most teams. It covers 80% of use cases with minimal operational burden. Go custom only when sub-millisecond latency or complex streaming transformations are hard requirements. ## Feast Feature Definition ### feature_store.yaml ```yaml project: my_ml_project registry: gs://my-bucket/feast/registry.pb provider: gcp online_store: type: redis connection_string: redis://10.0.0.5:6379 offline_store: type: bigquery entity_key_serialization_version: 2 ``` ### Featu
- Architecture Decision Table
- Feast Feature Definition
- featurestore.yaml
- Feature Definitions
- Point-in-Time Join
- How It Works
- Feast Point-in-Time Retrieval
- Online Serving Setup
- Materialization
- Online Retrieval for Inference
- Push-Based Streaming Features
- Gotchas and Anti-Patterns
- Training-Serving Skew
- Time-Travel Bugs
Materialize features from offline -> online store feast materialize-incremental $(date -u +"%Y-%m-%dT%H:%M:%S") Or in code
What does the feature-store-design skill do?
Online/offline feature serving, point-in-time correctness, Feast patterns, and feature computation design.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill feature-store-design --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.