Rerun

Sobre Rerun

Rerun is the unified data layer for physical AI. Physical AI teams use Rerun to ingest multi-rate, multimodal data from teleop, fleet logs, human-data rigs, and sim, visualize episodes in realtime, query and curate with dataframes or SQL, and stream dataset mixes directly into training. One columnar store underneath the whole pipeline. No export jobs, no stale copies. SDKs in Python, Rust, and C++. Built in Rust on column-chunk storage purpose-built for multi-rate physical data.