cs.DB · 2607.08319 Copy arXiv ID · Jul 9, 2026 Save GitLake: Git-for-data for the agentic lakehouse Authors: Weiming Sheng , Jinlang Wang , Manuel Barros , Aldrin Montana , Jacopo Tagliabue , Luca Bigon
Abstract We present GitLake, a Git-for-data design for an agent-first lakehouse. The system lifts single-table Iceberg snapshots into lakehouse-wide commits, branches, and merges, letting agents work on isolated branches while humans review and publish changes. Pipelines run on temporary branches and publish through a final merge, so all outputs become visible atomically or none do. Finally, we report production lessons as well as correctness insights from a preliminary Alloy model of our core abstractions.
Explore similar work May 31, 2026 · Nicole Rose Schneider, Davide Ghilardi, Giacomo Piccinini +1 Evaluation Harnesses Code Generation
Apr 16, 2026 · Shreesha G. Bhat, Tony Hong, Michael Noguera +2 Traditional Dataflow Analysis Data Lakes
May 31, 2026 · cs.AI J/K move · Enter open · S save
Nicole Rose Schneider, Davide Ghilardi, Giacomo Piccinini, Jacopo Tagliabue
Coding agents are becoming users of data infrastructure, but their success depends not only on model quality: it also depends on the skills and environment files that teach agents how to use a system. We study how to optimize these artifacts for agents operating on a branching lakehouse, Bauplan. In our setting, headless APIs and Git-like data primitives expose data workflows through code, branches, commits, and merges. Our central observation is that a branching lakehouse turns data-agent evaluation from an output-matching problem into a state-verification problem: agent-generated pipeline code induces concrete, inspectable lakehouse changes. We present a data-centric optimization pipeline that generates task-verifier pairs, executes candidate skills in isolated sandboxes, and scores trajectories using both trace-level signals and programmatic checks over lakehouse state. In a preliminary evaluation on hundreds of tasks, optimized skills improve held-out reward by up to 28.6%. These results suggest that write-path data workflows provide a useful substrate for optimizing agent skills beyond read-only tasks.