About
About Datahike
Datahike is an open-source immutable Datalog database. Related replikativ projects apply the same copy-on-write approach to columnar SQL, vector search and Lucene indexes. The work is maintained by an open-source community, with commercial support available for teams putting it into production.
Where Datahike comes from
The work started more than a decade ago with Votorola, a liquid-democracy project that needed a distributed data model. That led me to Clojure, persistent data structures and eventually Datahike. Its database values are immutable, historical states remain queryable and branches share unchanged storage. Collaborators including Konrad Kühne and the broader replikativ community helped shape what it is today.
Who is building this
Christian Weilbach, founder
I'm a researcher and systems builder. I started Datahike to explore databases that keep their history and remain useful as ordinary program values. My PhD at the University of British Columbia was on Structured Amortized Variational Inference: inference systems that accumulate evidence and maintain structured beliefs over time.
Before UBC, I completed a Master's at Heidelberg University on dictionary learning with Bayesian GANs and a Bachelor's project on online learning for Boltzmann machines using spiking-neuron models. I also studied philosophy and cultural anthropology, which still informs how I think about knowledge representation.
My research has appeared at ICML, NeurIPS, AISTATS and TMLR. I received a Google grant for work on Graphically Structured Diffusion Models and have spoken at Google DeepMind, OpenAI and MILA. I also co-organized the Clojure meetup in Mannheim-Heidelberg and the Machine Learning Meetup Rhein-Neckar.
Datahike connects those interests in a practical system: it accumulates facts, exposes its history and lets applications inspect earlier states with the same query engine.
Contributors
Many contributors, including our former team at Lambdaforge UG, shaped Datahike's early development. Their work added history indexes, time travel and temporal queries, bringing Datahike closer to Datomic's data model while keeping the implementation open source.
Today, Datahike is maintained independently and shaped by its broader open source community. Check out the full contributor list.
Join us
Build with us
Datahike is open source and welcomes collaborators. Current work includes documentation, integrations, storage backends and query-engine improvements. The issue tracker is the best place to find a concrete starting point.
Work with us
I'm building a company on top of Datahike. If you need help getting it into production, we can help with integration, custom development, and support contracts.
Email: contact@datahike.io