{"aiPlatform":"claude-code@2025.06","category":"data-analysis","commandName":"/data-driven-feature","content":"---\nname: Data-Driven Feature Development\ndescription: Build data-driven features with integrated pipelines and ML capabilities using specialized agents\nallowed_tools:\n  - memory          # For storing analysis and outputs from each agent\n  - sqlite          # For data analysis and pipeline requirements\n  - filesystem      # For implementing pipelines and ML code\ntags:\n  - data-science\n  - machine-learning\n  - pipeline\n  - analytics\n  - workflow\ncategory: development\nversion: 2.0.0\nauthor: AI Commands Team\n---\n\nBuild data-driven features with integrated pipelines and ML capabilities using specialized agents:\n\n[Extended thinking: This workflow orchestrates data scientists, data engineers, backend architects, and AI engineers to build features that leverage data pipelines, analytics, and machine learning. Each agent contributes their expertise to create a complete data-driven solution.]\n\n## Phase 1: Data Analysis and Design\n\n### 1. Data Requirements Analysis\n- Use Task tool with subagent_type=\"data-scientist\"\n- Prompt: \"Analyze data requirements for: $ARGUMENTS. Identify data sources, required transformations, analytics needs, and potential ML opportunities.\"\n- Output: Data analysis report, feature engineering requirements, ML feasibility\n\n### 2. Data Pipeline Architecture\n- Use Task tool with subagent_type=\"data-engineer\"\n- Prompt: \"Design data pipeline architecture for: $ARGUMENTS. Include ETL/ELT processes, data storage, streaming requirements, and integration with existing systems based on data scientist's analysis.\"\n- Output: Pipeline architecture, technology stack, data flow diagrams\n\n## Phase 2: Backend Integration\n\n### 3. API and Service Design\n- Use Task tool with subagent_type=\"backend-architect\"\n- Prompt: \"Design backend services to support data-driven feature: $ARGUMENTS. Include APIs for data ingestion, analytics endpoints, and ML model serving based on pipeline architecture.\"\n- Output: Service architecture, API contracts, integration patterns\n\n### 4. Database and Storage Design\n- Use Task tool with subagent_type=\"database-optimizer\"\n- Prompt: \"Design optimal database schema and storage strategy for: $ARGUMENTS. Consider both transactional and analytical workloads, time-series data, and ML feature stores.\"\n- Output: Database schemas, indexing strategies, storage recommendations\n\n## Phase 3: ML and AI Implementation\n\n### 5. ML Pipeline Development\n- Use Task tool with subagent_type=\"ml-engineer\"\n- Prompt: \"Implement ML pipeline for: $ARGUMENTS. Include feature engineering, model training, validation, and deployment based on data scientist's requirements.\"\n- Output: ML pipeline code, model artifacts, deployment strategy\n\n### 6. AI Integration\n- Use Task tool with subagent_type=\"ai-engineer\"\n- Prompt: \"Build AI-powered features for: $ARGUMENTS. Integrate LLMs, implement RAG if needed, and create intelligent automation based on ML engineer's models.\"\n- Output: AI integration code, prompt engineering, RAG implementation\n\n## Phase 4: Implementation and Optimization\n\n### 7. Data Pipeline Implementation\n- Use Task tool with subagent_type=\"data-engineer\"\n- Prompt: \"Implement production data pipelines for: $ARGUMENTS. Include real-time streaming, batch processing, and data quality monitoring based on all previous designs.\"\n- Output: Pipeline implementation, monitoring setup, data quality checks\n\n### 8. Performance Optimization\n- Use Task tool with subagent_type=\"performance-engineer\"\n- Prompt: \"Optimize data processing and model serving performance for: $ARGUMENTS. Focus on query optimization, caching strategies, and model inference speed.\"\n- Output: Performance improvements, caching layers, optimization report\n\n## Phase 5: Testing and Deployment\n\n### 9. Comprehensive Testing\n- Use Task tool with subagent_type=\"test-automator\"\n- Prompt: \"Create test suites for data pipelines and ML components: $ARGUMENTS. Include data validation tests, model performance tests, and integration tests.\"\n- Output: Test suites, data quality tests, ML monitoring tests\n\n### 10. Production Deployment\n- Use Task tool with subagent_type=\"deployment-engineer\"\n- Prompt: \"Deploy data-driven feature to production: $ARGUMENTS. Include pipeline orchestration, model deployment, monitoring, and rollback strategies.\"\n- Output: Deployment configurations, monitoring dashboards, operational runbooks\n\n## Coordination Notes\n- Data flow and requirements cascade from data scientists to engineers\n- ML models must integrate seamlessly with backend services\n- Performance considerations apply to both data processing and model serving\n- Maintain data lineage and versioning throughout the pipeline\n\nData-driven feature to build: $ARGUMENTS","contentHash":"a292c2346dc378837e42e2c937e50a95e4fa1b28d52a220cd1182212bca69eee","copies":1,"createdAt":"2025-08-12T16:09:34.356Z","description":"ML-powered features with data science subagents","github":{"repoUrl":"https://github.com/Commands-com/commands","lastSyncDirection":"from-github","metadata":{"importedFrom":"github_repository","repoPrivate":false,"repoDefaultBranch":"main","connectedAt":"2025-08-12T16:09:34.356Z"},"importedAt":"2025-08-12T16:09:34.356Z","lastSyncAt":"2025-08-17T17:57:49.063Z","fileMapping":{"license":null,"readme":null,"assets":[],"mainFile":"workflows/data-driven-feature.md"},"selectedCommand":"data-driven-feature","fileShas":{"mainFile":"e973ad566761019f3f64eb50b54f16e1f6f55f9a","yamlPath":"313647b1fb381389da33b7913e95baf617c4b392"},"branch":"main","connectionType":"commands_yaml","connected":true,"lastSyncCommit":"01591bc061d236bde47bf23b0f47e8afcf1a5144","importSource":"repository_import","installationId":"69232615","syncStatus":"synced"},"githubRepoUrl":"https://github.com/Commands-com/commands","id":"6eafa2bd-5394-44de-9fa5-7607160f6ef1","inputParameters":[{"defaultValue":"database","name":"data_source","options":["database","streaming","api","files","data-lake","hybrid"],"description":"Primary data source type","label":"Data Source Type","type":"select","required":false},{"name":"ml_use_case","options":["recommendation","classification","prediction","nlp","computer-vision","anomaly-detection"],"description":"Type of ML functionality","label":"ML Use Case","type":"select","required":false}],"instructions":"ML-powered features with data science subagents","lastCopied":"2026-01-11T05:07:09.239Z","likes":0,"mcp_search_content":"","organizationUsername":"commands-com","price":"free","search_content":"data driven feature ml-powered features with data science subagents /data-driven-feature data-analysis claude-code@2025.06","title":"Data Driven Feature","type":"command","updatedAt":"2025-08-17T17:57:49.063Z","userId":"W0V8NAw5AhWRwcuwSoFLOi1Yem83","visibility":"public","name":"data-driven-feature","userInteraction":{"userHasStarred":false}}