Natural Language Data Analysis
Asking a question of company data used to mean finding the right person and waiting for them to pull the numbers. This system answers in plain language, in seconds.
The problem
People need answers from company data every day and most cannot write SQL. Dashboards only answer the questions someone already thought to ask, so every other question means waiting for whoever can write the query.
What we built
You ask in plain language. The AI writes the database query, the database runs it and only the result comes back into the conversation - the AI never sees the raw data.
It runs on 70,000 rows across six business tables: budget, transactions, orders, customers, revenue targets and employees.
The part that makes it work is a context layer: a structured file that tells the AI what the data means - table relationships, naming conventions and business rules.
Stack: DuckDB · MCP · Claude · Python · CSV/Excel
What changed
Questions that used to wait for someone else now get answered in the conversation where they were asked.