Managed DuckDB for lakehouse analytics
DuckDB SQL on managed compute, open Parquet storage, federated catalogs, and MCP access for teams and agents.
DuckDB execution, without DuckDB ops
Workers run DuckDB SQL, cache hot data, connect external sources, and return results over SQL clients, APIs, and MCP. Storage stays in object storage. Compute scales independently.
Managed Workers
Run DuckDB SQL on Altertable-managed compute without operating infrastructure yourself.
DuckDB SQL
Standard DuckDB syntax and types across managed and external catalogs.
Parquet storage
Managed catalogs stored as Parquet in object storage.
Federated SQL
Join lakehouse tables with Postgres, warehouses, bucket tables, and Iceberg.
Hot-data caching
Local caching on Workers for faster repeat queries.
MCP for agents
OAuth-scoped agent access. No shared database credentials.
One runtime for teams and agents
Local DuckDB does not cover shared access, durable storage, federation, or governance. Altertable adds the operational layer so DuckDB-style SQL can run in production.
Shared lakehouse catalogs
Managed catalogs backed by buckets and Parquet files.
APIs and SQL clients
HTTP API, Arrow Flight SQL, and Postgres-compatible clients.
BI and dbt
Connect existing BI and transformation tools to the same runtime.
Self-hosted Workers
Run compute next to your buckets when locality or egress matters.
What you get
DuckDB speed with the ops layer production teams need.
Shared SQL environment
One shared runtime instead of laptop-local .duckdb files.
Less warehouse gravity
Run lakehouse workloads without routing every use case through a central warehouse.
Open Parquet storage
Data stays in portable object-storage formats, not a proprietary layer.
Agent-ready access
Agents query live data through MCP, not copied credentials.
Frequently asked questions
Is Altertable hosted DuckDB?
Altertable runs DuckDB SQL, but adds managed catalogs, object storage, federation, APIs, caching, and MCP access.
Can I query existing DuckDB files?
Yes. A .duckdb file in a connected bucket can be exposed as a read-only external catalog. Use Altertable catalogs when you want Altertable to own writes and retention.
Where does compute run?
On Altertable-managed Workers, or on self-hosted Workers you deploy in your own environment.
Does it replace my warehouse?
Yes. Altertable can be your primary analytical runtime for interactive SQL, Parquet storage, dashboards, and agent access. External catalogs let you query existing sources during migration or alongside the lakehouse.
Explore the rest of the platform
More workloads on the same engine, governance, and pricing.
AI Context
Give Claude, GPT, and other agents governed access to trusted data, metrics, and business context. Agents work from the same definitions and permissions as your team.
Tools & Integrations
Plug Altertable into BI, notebooks, SQL clients, and agents through open standards like Iceberg, Parquet, the Postgres wire, Arrow Flight, and MCP. Adopt it alongside your stack instead of replacing it.
Run DuckDB SQL on a managed lakehouse
DuckDB workers on open formats, federated SQL across your existing systems,
and an MCP server for agents — at flat monthly pricing.