Rerun

About Rerun

Rerun is an open-source, high-performance visualization and data infrastructure platform designed for spatial and embodied AI applications.

It provides SDKs in C++, Python, and Rust, enabling engineers and researchers to easily stream, log, and visualize complex multimodal data in real-time or from recordings.

Its robust data model, based on a time-aware Entity Component System, simplifies common workflows like modeling, debugging, and analyzing robotic systems or AI prototypes.

Rerun supports interactive, multi-timeline data visualization both natively and in-browser, with fast querying capabilities for extracting time-aligned datasets.

It integrates seamlessly into existing workflows, allowing embedding in applications or notebooks, and offers features like data streaming, logging, visualization, and debugging at scale.

Additionally, Rerun is used by projects such as LeRobot, Aria Dataset Explorer, and Brush, showcasing its flexibility across various research and industrial applications.

It is under active development, with a focus on scalable data management, entity filtering, and partial updates, supported by $17 million in seed funding to build a comprehensive data stack for Physical AI..

Smart Features

• Open-source platform supporting C++, Python, and Rust SDKs for versatile integration.
• Fast, interactive visualization for live and recorded spatial and embodied AI data.
• Seamless multi-timeline data streaming, logging, and serialization management.
• Robust data modeling with a time-aware Entity Component System for simplicity and flexibility.
• Embedded visualization capabilities in web, native, and notebook environments.
• Powerful query APIs for extracting and analyzing time-aligned datasets.
• Supports large-scale data ingestion, storage, and management with built-in debugging tools.
• Used by leading AI and robotics projects for visualization and training.
• Open-source licensing under MIT and Apache 2 for community collaboration.
• Active development with features like entity filtering, partial updates, and backwards compatibility.

Use Cases & Applications

• Real-time visualization and debugging of robotic and AI systems.
• Data logging and analysis for training deep learning models in robotics.
• Visualization of spatial and embodied AI data for research projects.
• Integrated visualization during model development and prototype testing.
• Analysis and extraction of signals from complex multimodal logs.
• Embedding visualization tools into custom robotics or AI applications.
• Supporting datasets exploration such as Meta’s Aria Dataset and Deepmind’s Brush engine.

Who is it for?

• AI researchers working on spatial and embodied intelligence.
• Robotics engineers developing perception and control systems.
• Data scientists analyzing multimodal robotic logs.
• Software developers integrating visualization into AI tools.
• Academic and industry labs focusing on Physical AI and computer vision.

Business Opportunities in Rerun

Leverage Rerun’s powerful visualization and data management capabilities to develop custom AI tools, training pipelines, or debugging solutions for robotics startups or research labs.

By building specialized applications or plugins around Rerun, entrepreneurs can offer tailored visualization services, data analysis tools, or training modules.

The open-source nature allows for rapid customization and deployment, providing a competitive edge in the growing Physical AI market.

Additionally, creating niche integrations or consulting services for companies adopting Rerun can generate revenue while supporting innovative AI projects..