Scaling AI for complex biological systems means scaling its iterative cycle: laboratory results inform models, models design the next round of experiments. While leading ML and Comp Bio teams in the industry we experienced firsthand that — beyond data generation itself — the primary bottleneck was closing the feedback loop across heterogeneous modalities at scale. We needed a way to query, trace, and validate complex biological datasets with context and memory to effectively learn from mistakes and train the next generation of models.

We founded Lamin in 2022 to build an open-source, zero-lock-in solution to this problem. We started with a "git for R&D data" that enabled traceability & versioning alongside a biological data catalog. Today, LaminDB has evolved into a lineage-native lakehouse: a programmable context and memory layer for biological R&D that scales—just like git—from personal projects to pharma-scale enterprise deployments.

We are proud to work with thousands of scientists across academia, biotech, and global pharma. Together, we are building a more reliable foundation for data-driven research at scale.

Headquartered in Munich and New York City, we are always looking for exceptional people to join our mission — please reach out!