AI Query Handler & Operations Workbench
A React and n8n interface that turns natural-language operating questions into grounded Airtable answers and action drafts.
The workbench explored how a role-specific AI interface can sit above existing operational data without replacing the underlying systems of record.
Users can ask questions such as which leads arrived last week in a specific segment. The workflow converts the request into a bounded data operation, retrieves matching records and returns a concise answer.
Two intake forms feed Google Sheets and then Airtable, giving the workbench a consistent data foundation for lead and KPI questions.
Action-oriented responses can prepare email drafts while keeping the user in control of the final message and recipient.
Operational data existed across forms, sheets and Airtable, but retrieving answers or preparing actions still required manual filtering and context switching.
A conversational React interface routes natural-language questions through n8n, queries structured Airtable data, returns grounded answers, and can prepare email drafts and operational summaries.
Outcomes without invented claims
- 01Combined natural-language retrieval, KPI context and email drafting in one interface.
- 02Kept structured Airtable data as the source of truth rather than relying on ungrounded model memory.
- 03Demonstrated a reusable pattern for role-specific operations workbenches.
What makes this publishable
- React interface
- n8n query workflow
- Airtable schema
- Form-to-Sheets-to-Airtable ingestion flow
- React
The project shows how AI becomes useful when it is constrained by real operational data and designed around the decisions a specific user needs to make.