synthdata-prompt-builder

Planning a synthetic data workflow involves translating domain requirements into a coherent, end-to-end plan that maps raw tables, derived tables, and reference datasets to concrete generation and computation commands.

Generating synthetic data for healthcare risk scoring or fraud detection is best handled by matching user goals to domain templates, which produces a production-ready plan linking raw and derived tables to concrete commands.

Yes, you can use templates to map raw and derived tables for IoT sensor dashboards by applying domain classification to your requirements and generating a step-by-step plan with concrete prompts.

Synthetic data generation requires defining both generation and computation steps to ensure the workflow produces raw tables, derived tables, references, and assets for your specific domain.

To create an end-to-end synthetic data plan, you need to provide a description of your domain and requirements, which the system uses to enforce domain classification and match your goals to existing templates.

You should avoid a template-based approach for synthetic data generation when your scenario falls outside standard use cases like healthcare risk or ecommerce fraud, as the workflow relies on matching existing domain templates to produce plans.