CedarDB brings unified database processing to the agentic AI era

Born as a spinout from the Technical University of Munich, CedarDB leverages the architecture of the Umbra research project to optimize performance on modern hardware. CEO Moritz Sichert reports that the engine delivers faster processing speeds without requiring additional memory or CPU resources. Crucially, the system maintains compatibility with PostgreSQL, allowing companies to integrate the technology without rebuilding their existing application layers.

Following a €5.3 million seed round, the startup is now expanding its reach beyond on-premise deployments. The team officially launched limited availability for CedarDB Cloud at TechCrunch Disrupt, targeting a broader user base that extends past large enterprise clients. This move places the 14-person team in direct competition with major players like Databricks, which recently introduced its own LTAP architecture to unify transaction and analytical workloads.

While the market for hybrid databases remains competitive, Sichert views the industry's shift toward unified storage as validation of his team's decade-long development focus. With the cloud offering now live, the company plans to scale its operations to meet anticipated demand, setting the stage for future fundraising to support its growth.