Upload a file, connect a database, or just start talking. Datrov handles the analysis, training, and deployment.
Datasets, charts, and trained models land in your library automatically, whether they came from a single chat or a whole afternoon of iterating.
Datrov writes the SQL, runs it against your live database, and explains the result in plain language.
Training a model is only half the job. Datrov deploys it automatically, gives it its own API key, and keeps track of every version in your library so you can call it from Python, JavaScript, or curl whenever you need it.
CSV, Excel, JSON, Parquet, SQLite files, or a live PostgreSQL or MySQL database. Upload a file or paste in connection details, and Datrov reads the schema automatically.
No. Describe what you want in plain language. Datrov writes and runs the Python behind the scenes, and shows you exactly what it ran.
Your files and database connections stay scoped to your account and session. Nothing is shared across workspaces.
Yes. Every trained model can be deployed as a live endpoint with its own API key, callable from Python, JavaScript, or curl.
You can top up with a one-time credit pack, or upgrade to a plan with a larger monthly allowance. Nothing gets deleted.