Samskara is now in early access

Over the last several months we’ve built a Python notebook IDE, a SQL Workbench, a cron-based scheduler, SQL-driven dashboards, a DuckDB + Iceberg lakehouse we call ArrowLake, an AI copilot that actually understands our own APIs, governance and RBAC, Git integration, and a Unified Data Explorer that ties it all together. We’ve run all of it end-to-end on real projects — weather data, public health datasets, the Northwind sample database rebuilt as a real pipeline.

That’s enough to stop calling it an idea.

Samskara is the self-hosted data engineering platform for teams that have outgrown scripts but don’t need the complexity and cost of a large cloud data platform.

If you’re a 20–200 person team and you’ve ever said any of the following, this is for you:

• “I don’t want to maintain Spark clusters.”
• “I don’t need Databricks for a company our size.”
• “My team just wants notebooks and SQL.”
• “I need governance, but I don’t need an enterprise sales cycle to get it.”

We’re opening up early access starting today. The Community Edition isn’t public yet — we’re rolling it out to teams directly first. If that’s you, book a demo and we’ll get you running against your own data.

Over the coming weeks we’ll be writing about the decisions behind the platform — starting with why we built ArrowLake instead of just picking a warehouse off the shelf.