Most of us are still using database tools that were designed for a different era.
TablePlus. DBeaver. Beekeeper Studio. Great tools. I've used all of them. But they share the same mental model: connect to a database, run a query, browse some rows. Repeat.
That model made sense when your data lived in a single Postgres or MySQL instance. It doesn't map well to how modern data work actually happens.
Today you're querying Parquet files directly. Local CSVs. S3 buckets. Remote HTTP endpoints. You're writing Python alongside SQL. You're building analyses that need version control, reproducibility, and AI assistance — not just a results grid.
The old tools bolted some of this on. DuckDB as one connection type among twenty. An "AI shell" that suggests SQL. Export to CSV. Done.
That's not wrong. It's just not enough.
I built Ruddy because DuckDB deserves to be the center of the experience, not a checkbox.
Ruddy is a macOS-native IDE where DuckDB isn't a plugin — it's the engine everything is built on.
What that looks like in practice:
No WASM, no Electron
The native DuckDB binary runs directly on your Mac. Full performance, full extension support, no browser sandbox overhead.
Notebooks are first-class
Powered by Marimo, with Python + SQL in the same environment and installable native libraries. Not a cell runner bolted on top — a real analysis environment.
An AI agent that actually executes
Ruddynie doesn't just suggest SQL. She runs queries, builds notebooks, and connects to MCP servers. You bring your own model — Claude, Gemini, OpenAI, OpenRouter.
Spatial out of the box
Geospatial data renders as maps without configuration. DuckDB's spatial extension is fully exposed.
Git built in
Your SQL and notebooks are code. They live in your repo, with version control inside the editor.
Visual query builder via ERD
Click together joins from the schema diagram. Useful for exploration and for teammates who don't live in SQL.
The old tools aren't going away, and they shouldn't. If you need a universal client for Postgres, MySQL, and fifteen other databases, they serve that well.
But if DuckDB is central to how you work — local files, analytical queries, Python notebooks, AI-assisted analysis — you deserve a tool built for that workflow from the ground up.
That's Ruddy.
→ ruddy.pro — macOS, 7-day free trial. Early-access pricing, $15 one-time or $1.25/month.
Top comments (0)