Project Demo · GenAI Analytics Agent

Natural Language
to SQL

Ask a business question in plain English. The system retrieves the relevant schema, business definitions and verified examples from a pgvector knowledge base, writes PostgreSQL with an open-weight model, validates it, and runs it read-only against 40,000+ rows of synthetic online-retail data: orders, customers, products, deliveries and inventory.

Schema

order_idcustomer_idproduct_idorder_datestatustotal_amountchannel
customer_idsegmentlifetime_valueacquisition_dateregionloyalty_tier
  1. Retrieve
  2. Generate
  3. Validate
  4. Run

// Ask a question below or click ⊞ to preview a table

Open-weight models via Groq · RAG over Supabase pgvector · Click ⊞ to preview a table