Natural Language Data Analysis
Querying business data used to mean finding the right person, waiting for them to pull the numbers, and interpreting the result. We built a system that answers data questions in plain language, in seconds.
The problem
Business users need answers from company data every day. Most can’t write SQL. Dashboards only answer questions someone already thought to ask. Every ad hoc question means waiting for the data team.
What we built
A system where you ask questions in plain language. The AI generates the database query, the database executes it, and only the results come back to your conversation. The AI never sees the raw data.
Currently running with 70,000 rows across 6 business data tables: budget, transactions, orders, customers, revenue targets, and employees.
The key is a context layer - a structured file that tells the AI what the data means: table relationships, naming conventions, and business logic rules.
Stack: DuckDB · MCP · Claude · Python · CSV/Excel
What changed
Questions that previously required finding the right person, waiting for them to pull the data, and interpreting the result now take seconds.