feast-feature-store

name	feast-feature-store
description	Feature store management skill for online/offline feature serving, feature registration, and training-serving consistency.
allowed-tools	["Read","Write","Bash","Glob","Grep"]
graph	{"domains":["domain:data-science"],"specializations":["specialization:data-science-ml"],"skillAreas":["skill-area:feature-engineering","skill-area:feature-engineering-pipelines"],"roles":["role:ml-engineer","role:data-engineer"],"workflows":["workflow:ml-model-lifecycle"],"topics":["topic:data-mesh"]}

feast-feature-store

Overview

Feature store management skill using Feast for online/offline feature serving, feature registration, and ensuring training-serving consistency in ML systems.

Capabilities

• Feature definition and registration
• Online feature serving setup
• Offline feature retrieval for training
• Point-in-time correctness validation
• Feature freshness monitoring
• Entity management
• Feature view creation and management
• Materialization scheduling

Target Processes

• Feature Store Implementation and Management
• Feature Engineering Design and Implementation
• Model Training Pipeline

Tools and Libraries

• Feast
• Redis (online store)
• PostgreSQL/BigQuery/Snowflake (offline store)
• Parquet files

Input Schema

{
  "type": "object",
  "required": ["action"],
  "properties": {
    "action": {
      "type": "string",
      "enum": ["apply", "materialize", "get-online", "get-historical", "list", "teardown"],
      "description": "Feast action to perform"
    },
    "featureRepo": {
      "type": "string",
      "description": "Path to feature repository"
    },
    "features": {
       
          
       
    
     
       
       
    
     
       
       
            
            
      
    
  

Output Schema

{
  "type": "object",
  "required": ["status", "action"],
  "properties": {
    "status": {
      "type": "string",
      "enum": ["success", "error"]
    },
    "action": {
      "type": "string"
    },
    "features": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "name":    
              
              
              
        
      
    
     
       
       
            
            
      
    
     
       
       
    
  

Usage Example

{
  kind: 'skill',
  title: 'Retrieve training features',
  skill: {
    name: 'feast-feature-store',
    context: {
      action: 'get-historical',
      featureRepo: 'feature_repo/',
      features: ['user_features:age', 'user_features:tenure', 'transaction_features:avg_amount'],
      entityDf: 'data/training_entities.parquet'
    }
  }
}