Five computer scientists who spent years tinkering with database architecture in a Technical University of Munich lab have emerged with €5.3 million (approximately $6 million) in seed funding and a provocative claim: their PostgreSQL-compatible engine, they say, runs some workloads 100 times faster than the original.
CedarDB announced the financing on May 14, led by Amplify Partners and backed by a roster of angel investors that reads like a who's who of data infrastructure. Hanno Renner, the chief executive of HR software giant Personio, joined the round. So did Jordan Tigani, who runs MotherDuck, and Wes McKinney, the creator of the ubiquitous pandas library and Apache Arrow who now operates Composed Ventures. Three additional angels participated but were not named.
The startup is chasing a technical problem that has bedeviled database engineers for decades. Traditional systems force companies to choose: optimize for rapid transactions, the lifeblood of e-commerce checkouts and financial trades, or build separate infrastructure for analytical queries that crunch historical data. CedarDB's founders argue their engine handles both in a single system, eliminating the complexity and cost of maintaining parallel databases.
Moritz Sichert, the company's chief executive, has a background that straddles business and technology. He previously served in a finance role before earning his doctorate at TUM, where he and his four co-founders—Lukas Vogel, Christian Winter, Dominik Durner and Philipp Fent—worked on a research project called Umbra. All five hold PhDs. Vogel did a stint at Salesforce, Winter at Oracle Labs, Durner at both SAP and Microsoft. They spun out CedarDB in 2024 with backing from their former professors, Thomas Neumann and Alfons Kemper, who now advise the company.
The seed capital will fund customer acquisition and a push into U.S. markets, according to honert, the Munich law firm that advised on the deal. The firm disclosed the allocation in mid-June, noting that CedarDB had wrapped initial pilot programs and was targeting financial services firms, including banks and cryptocurrency exchanges.
CedarDB released two versions of its database on the day it announced the funding. A Community Edition, free but capped at 64 gigabytes of compressed data, aims to attract developers. An on-premises Enterprise Edition targets larger organizations unwilling to move sensitive workloads to the cloud. Christian Winter, one of the co-founders, detailed the releases in a company blog post at the time.
The startup later launched a fully managed cloud service. Configurations range from modest two-virtual-CPU instances to brawny setups packing 192 vCPUs and 768 gigabytes of memory. At rollout, the service entered limited availability with a free trial for certain tiers. As of October 2026, the company employed 14 people, according to industry coverage.
CedarDB's timing may prove fortuitous. The database market has shifted as companies grapple with new demands from artificial intelligence workloads, particularly agentic systems that rapidly toggle between writing data and analyzing it. Databricks unveiled an initiative around this challenge last summer, and competitors including SingleStore and Google's AlloyDB for PostgreSQL have pushed similar unified engines. "The approach has gained relevance with the rise of agentic AI," Sichert told one tech publication.
The company has steadily added features: support for newer PostgreSQL protocols, compatibility with pg_stat system tables used for monitoring, and various query optimizations. Its performance claims are eye-catching, though CedarDB has not disclosed the names of paying enterprise customers. The startup does appear on ClickBench, a public leaderboard where analytical database systems race to execute a standardized suite of queries. There it sits among a mix of OLAP and hybrid transactional-analytical systems.
Amplify Partners, in a blog post last year, called CedarDB "the fastest database you've never heard of" when disclosing its investment. The venture firm did not respond to a request for comment on valuation or deal structure, which CedarDB declined to share.
Whether the Munich team can convert technical speed into commercial traction remains an open question. The database landscape is littered with elegant systems that failed to unseat incumbents. But the pedigree is undeniable, the market timing looks plausible, and the investor lineup suggests at least a few industry insiders think these five PhDs might be onto something.
