Analytics without the SQL bottleneck — or the data leak.
Ask your database a question in plain English (or Vietnamese). A local model writes the SQL, executes it read-only inside your perimeter, and explains the result — every query logged, no numbers leaving your network.
Start with a $999 pilot — limited to our first 5 customers (then $1,999). No production data leaves your network — or you don't continue.
What we deploy
- Read-only database access — the model can never modify data
- Every generated query logged for audit
- Works against your existing warehouse, database, or spreadsheets
- Bilingual English / Vietnamese question support
- The answer arrives with its SQL and a chart, so anyone can check the logic
See it in action
What it looks like in practice
The same architecture, applied across very different teams — your industry is just the data source.
A CPA firm asks: "Which clients' receivables aged past 60 days last quarter?" — no analyst queue, no exported spreadsheets.
An e-commerce operator asks: "What was my best-margin category by month this year?" straight against the orders database.
A property manager asks: "Which units had the highest maintenance spend per square foot?"
A clinic manager asks: "How many no-shows did we have by provider last month?" without touching raw tables.
A salon group asks: "Which technicians had the highest rebooking rate last quarter, by location?" straight from the POS export.
An owner asks in Vietnamese — "Doanh thu tháng này so với cùng kỳ năm ngoái thế nào?" (how does this month's revenue compare with the same month last year?) — and gets the answer in Vietnamese.
See it live: Ask Your Books
A working proof of concept: ask plain-English (or Vietnamese) questions about a sample books database and watch a local model write the SQL, run it read-only, and explain the answer — without a single byte leaving the machine.
What the 30-day pilot looks like
One workflow, your real documents, and a measured result before you commit to anything larger.
Week 1: we map the schema with you and collect 30 questions your team actually asks, with the answers you already trust.
Weeks 2–3: read-only connection, schema glossary and local model deployed; your team starts asking.
Week 4: an accuracy report on those 30 questions — every generated query shown — and a go/no-go on wider rollout.
Common questions
Do I need to know SQL to use it?
No — that is the point. You ask in plain English or Vietnamese and get the answer plus a chart. The generated SQL is shown alongside so anyone technical on your team can check the logic.
Can it change or delete our data?
No. It connects with read-only credentials, so the worst case is a query that returns nothing useful. Write access is never part of the deployment.
How do we know the numbers are right?
Because the numbers come from a SQL query against your database, not from a language model's memory. The model writes the query; your database computes the answer. You can read the query, run it yourself, and compare.
Which databases does it work with?
PostgreSQL, MySQL, SQL Server, and most systems with a standard SQL interface. We map your schema during setup so the model knows what your table and column names actually mean.
Does our data leave the network?
Not in an on-premise deployment. The model, the query engine, and the database all sit inside your perimeter, and every query is logged for audit.
What about spreadsheets that aren't in a database?
We load them into a local database during setup, and recurring exports can be refreshed on a schedule. Questions then run against that copy, so the original files are never touched.
Related use cases
Proven in regulated industries
We've packaged this capability for compliance-heavy fields — see how it lands in yours.
Want this running on your hardware?
Book a 30-minute discovery call. We'll scope a $999 pilot — limited to our first 5 customers (then $1,999) — no data leaves your network, or you don't continue.