Open source data platform

Build data workflows
you can maintain.

curl -LsSf getrenart.com/install.sh | sh

Build your first pipeline Star us on GitHub

Public alpha Apache 2.0 · Runs locally

Renart is an open source data platform for moving, transforming and analyzing data across your data stack.

Build and run SQL and Python pipelines alongside notebooks and dashboards in one visual workspace, with type checking, scheduling and everything defined as code in Git.

Take a look inside

Illustrated walkthrough64 seconds · no sound

Jump to

Renart is in public alpha. Expect rough edges and changes before the first stable release. What to expect

The entire data workflow

From getting data in
to putting it to work.

Bring the data together.

Load from databases, files, object storage and HTTP APIs. Move data into and out of supported warehouses as part of the pipeline.

Data loading

Build with SQL and Python.

Develop transformations with type checking and intelligent completion. Connect the assets, inspect the results and schedule the work.

Pipeline development

Make the results useful.

Explore data in notebooks and build dashboards and reports. Keep the analysis and its definitions alongside the pipelines in Git.

Notebooks and analysis

Built for the next change

Make changes
with context.

A pipeline is easier to maintain when the code, dependencies and results are in view together.

Explore the asset editor

SQL completion in Renart suggesting customer columns and their types from an upstream asset — Completion uses the tables and columns your pipeline knows about.

Catch mistakes while you edit.

Get SQL and Python diagnostics and completion in the editor. Check known columns and types before running against your database.

See what a change affects.

Follow dependencies on the canvas. Staleness highlights changed definitions and downstream assets that may need rebuilding.

Review before you run.

Preview SQL results with Inspect, review the file diff and choose what to materialize. Scheduled runs use a deployed snapshot.

One workspace, shared definitions

Build, explore and run
in one workspace.

Move between pipelines, notebooks and scheduled runs with the same assets, schemas and dependencies.

Load

Bring data into the workflow.

Choose a source, destination and load mode in the workspace. Make the transfer part of the pipeline, alongside its transformations.

Databases, files and object storage

Read the docs

Renart Load asset editor with source, destination and loading options — Actual Renart interface Open image to see the details

Local first. Defined in Git.

Your infrastructure.
Your repository.

Work with the stack you have.

Start locally with DuckDB or connect to your databases and warehouses. Bring in files, object storage and HTTP API data where the workflow needs it.

DuckDB
Postgres
ClickHouse
StarRocks
Trino
Snowflake
BigQuery
Redshift
Databricks
S3 / GCS
HTTP APIs

Connections and environments

Keep the work reviewable.

Pipelines, notebooks, dashboards, reports and schedule definitions are plain files. Visual edits become changes you can review, commit and revert in Git.

How Renart stores your work

Run it in your environment.

No Renart account or hosted control plane is required. You choose the systems Renart connects to. Credentials and machine-local run history stay separate from authored definitions.

Security and privacy

Start with a local pipeline

Put Renart to work.

Open a Git repository, build a small pipeline and see how the workflow fits.

Follow the quickstart

Install Renartcurl -LsSf getrenart.com/install.sh | shInstallation options for Linux, macOS and Windows

Build data workflows you can maintain.

From getting data in to putting it to work.