Why Kvatch
Data federation without the ETL tax. Query across PostgreSQL, SQLite, Google Sheets, and flat files as if they were one database — no pipelines to maintain.
Most teams have data scattered across a production database, a handful of spreadsheets, some CSV exports, and a few internal APIs. Answering a single business question often means exporting everything into one place, writing throwaway join logic, and hoping the numbers still line up. That process is slow, repetitive, and error-prone — so we built a CLI that treats every one of those sources as a table you can query and join directly.
Kvatch started as a command-line tool because that's where data engineers and developers already work. The CLI is free and open source under Apache 2.0, and will stay that way. The Runtime — a self-hostable API and orchestration layer — is available now and open source too. Kvatch Portal, a point-and-click interface for business users, is in alpha and will be the eventual paid tier for teams that want a hosted, collaborative experience.
We're a small, early-stage team building in the open. If you hit rough edges, that's useful signal — tell us on GitHub.
Open by default
The CLI and Runtime are Apache 2.0 and always will be. No bait-and-switch licensing.
Local-first
Your data stays on your machine unless you choose otherwise. No forced cloud round-trips.
No lock-in
Plain SQL and YAML config. Swap Kvatch out any time without rewriting your pipelines.