Flarka is an open-source data platform you can deploy in your own cloud — Azure, AWS, Hetzner, OVH — or on-premise. ETL pipelines, data catalog, notebooks. No Databricks bill. No American cloud dependency.
Powered entirely by open source
What you get
Everything Databricks offers, minus the SaaS pricing and the data leaving your infrastructure.
Run SQL transformations directly on your cloud storage — Azure Blob, S3, GCS. No clusters to manage. Processes tens of GBs on a single pod.
Open data catalog compatible with the Delta standard. Browse tables, track lineage, manage schemas — all through a clean REST API.
Argo Workflows handles your pipeline scheduling. Run jobs on a cron, trigger on events, monitor failures — GitOps native.
Each data engineer gets their own notebook environment with DuckDB pre-configured and direct access to your catalog tables.
Authentik provides single sign-on across the entire platform. One login for JupyterHub, workflows, catalog, and storage console.
Deploy on any Kubernetes cluster. Data never leaves your infrastructure. GDPR-compliant by design — we manage the platform, not your data.
vs Databricks
Databricks charges per DBU. Flarka charges a flat management fee. Your compute costs stay in your cloud account.
Pricing
You pay for platform management. Your compute costs go directly to your cloud provider.
Early access
Leave your email and we'll set up a 30-minute demo on your own infrastructure.