Databend
An open-source, Rust-built data warehouse unifying large-scale analytics, vector search and full-text search with sandboxed Python UDFs for running agent logic through SQL.
Databend is an open-source enterprise data warehouse built in Rust, unifying large-scale analytics, vector search, full-text search and automatic schema evolution in one engine. It runs on object storage such as S3, Azure Blob Storage or Google Cloud Storage, with elastic, cloud-native compute.
A sandboxed user-defined function lets a query define agent logic, such as LLM calls, tool use or reasoning, in Python, which SQL can then orchestrate and combine with transactions and Git-like branching for safe experimentation on production data snapshots. Its architecture separates a control plane that handles scheduling, permissions and sandbox lifecycle from an execution plane that issues requests over Arrow Flight and a compute plane of isolated sandbox workers that run the agent logic.
It can run as a managed cloud service, installed locally through a Python driver for development, or run as a full warehouse through Docker.