Amazon Acquires DuckLabs and Integrates DuckDB in Amazon Aurora PostgreSQL

Amazon has acquired DuckLabs, the Amsterdam-based company behind the analytical database DuckDB, and launched its first native integration inside Amazon Aurora PostgreSQL. Co-founders Hannes Mühleisen and Mark Raasveldt will continue leading the team and technical direction under Amazon Web Services. As covered in InfoQ's report, DuckDB, DuckLake, and Quack remain free and open-source under the MIT licence, governed by the non-profit DuckDB Foundation.

This report details the acquisition structure, the technical collaboration leading to the deal, and the new query capabilities added to Amazon Aurora PostgreSQL. It is intended for database administrators, data engineers, and cloud architects assessing how embedded analytics eliminate reverse ETL pipelines, streamline sub-terabyte analytical queries, and support agentic AI workloads across Amazon S3 object storage.

The deal behind Amazon DuckLabs and Integrates DuckDB

Amazon finalized the acquisition of DuckLabs, integrating the Amsterdam-based engineering organization into Amazon Web Services. DuckDB co-founders Hannes Mühleisen and Mark Raasveldt confirmed they will continue heading the engineering team and steering technical choices for the open-source database inside AWS. In their announcement, the founders stated that joining AWS provides the resources and reach to expand the technology to more organizations and pursue large-scale ideas difficult to achieve independently. DuckDB and Duck Stack projects, such as DuckLake and Quack, remain MIT-licensed open source under DuckDB Foundation governance.

The transaction resulted from roughly two years of collaboration between DuckLabs and AWS engineering teams. Andy Warfield, distinguished engineer at AWS, noted on Werner Vogels's blog that initial discussions began when the Amazon S3 team developed S3 Tables. Over time, the teams explored AWS Lambda as a matching runtime primitive for launching DuckDB queries rapidly across cloud environments. They also smoothed integrations between DuckDB and AWS database engines through DuckDB's evolving ATTACH and CONNECT commands, ensuring alignment before deciding to bring the DuckLabs team into AWS.

Why Amazon DuckLabs and Integrates DuckDB raised now

AWS targeted DuckLabs because modern analytics workloads increasingly concentrate on compact datasets. Mai-Lan Tomsen Bukovec, technology vice president at AWS, explained that DuckDB solves performance bottlenecks for over 90% of global data queries running on one terabyte or less for analysis and dashboarding. She emphasized that DuckDB's execution speed is particularly relevant for agentic AI workloads. AI agents routinely inspect underlying data structures and trigger exploratory queries before deciding subsequent actions, making DuckDB's lightweight, in-process analytical design naturally optimized for autonomous AI agent pipelines.

For DuckLabs, joining AWS provides engineering bandwidth and infrastructure distribution that an independent open-source venture could not match alone. Co-founders Mühleisen and Raasveldt recognized that AWS offers the scale necessary to deploy DuckDB across thousands of enterprise environments. Andy Warfield explained that Amazon wanted to give the co-founders the broadest opportunity to set the technical agenda for DuckDB and cloud analytics delivery. Rather than operating as an external database vendor, DuckLabs gained direct access to Amazon's core database ecosystem, enabling deep native integrations into production services.

What the Amazon DuckLabs and Integrates DuckDB money

The immediate product deliverable from the acquisition is a native DuckDB integration inside Amazon Aurora PostgreSQL. Esra Kayabali, principal solutions architect at AWS, explained that this capability simplifies architectures combining operational records with historical logs. Previously, applications querying live Aurora tables alongside historical data in Amazon S3 relied on reverse ETL pipelines that duplicated records, increased infrastructure costs, and demanded continuous engineering work to maintain synchronization. The integration removes those external data movement pipelines by executing analytical operations directly inside the Aurora database instance.

Aurora PostgreSQL users can execute a single SQL query joining live operational data with historical records across Apache Iceberg tables managed through AWS Glue Data Catalog, alongside Parquet and Iceberg datasets stored in Amazon S3 and S3 Tables. Processing remains within Amazon Aurora, eliminating network hops and extra ETL configurations. The feature is available across all commercial AWS regions and AWS GovCloud (US) at no additional charge, requiring customers to pay solely for incremental Aurora compute capacity and standard Amazon S3 requests generated by their queries.

Competitive context for Amazon DuckLabs and Integrates DuckDB

Integrating DuckDB into Aurora PostgreSQL changes how cloud relational engines address hybrid transactional and analytical processing. Traditionally, combining live operational tables with cold cloud storage required separate data lakehouses or dedicated query engines. By enabling Aurora to query Parquet and Iceberg tables directly via DuckDB, AWS allows engineering teams to analyze sub-terabyte datasets without provisioning dedicated external analytical clusters. This design directly challenges external analytics platforms by keeping analytical query processing inside the operational database engine while relying on standard Amazon S3 and S3 Tables for historical storage.

The transaction also gives AWS an advantage in developer tooling for emerging AI architectures. Because AI agents rely on quick exploratory queries across small datasets, embedding an optimized analytical database provides low-latency execution without managing separate infrastructure. Andy Warfield emphasized that AWS sought to provide the DuckLabs creators with a platform to define cloud analytics delivery. By integrating DuckDB with core primitives like Amazon S3 Tables and AWS Lambda, Amazon creates a unified environment that competes against standalone analytical engines and alternative cloud data platforms.

Open questions on Amazon DuckLabs and Integrates DuckDB

The community response to the acquisition has combined support with questions regarding open-source independence under a major cloud vendor. On Reddit, user adappergentlefolk noted satisfaction that the founders found a path to monetize their work while keeping code under foundation governance, but voiced concern that the acquisition might slow DuckLabs' historical pace of innovation. Although AWS confirmed that DuckDB and the Duck Stack remain under the MIT licence supervised by the non-profit DuckDB Foundation, developers will observe whether AWS prioritizes proprietary integrations over community-led enhancements.

Engineers also face architectural questions regarding resource contention when running analytical queries directly inside operational Aurora databases. Because DuckDB executes within Aurora rather than on separate clusters, database teams must monitor compute and memory allocation to prevent analytics from degrading transactional throughput. Teams will evaluate performance when running complex queries across Apache Iceberg tables in Amazon S3 and AWS Glue Data Catalog. Additionally, whether AWS will integrate other Duck Stack components, including DuckLake and Quack, into additional managed database services remains an open question for enterprise architects.