Automated pattern discovery engine for tabular data, built for science and AI agents
Visit DiscoDisco is an automated scientific discovery engine by Leap Laboratories that finds patterns in tabular data that humans and AI agents typically miss. It fits neural networks to uploaded datasets, applies interpretability methods to extract learned patterns, validates all findings on hold-out data, and contextualizes results against existing literature. Users receive a ranked list of statistically significant patterns complete with p-values, effect sizes, visualizations, and citations. It is accessible via a web interface, Python SDK, REST API, and MCP server, enabling fully programmatic use by AI agents.
Visit Disco's official website for product details and getting started.
Comprehensive guides on how to use Disco effectively, including setup and features.
Insights and updates about Disco, including case studies and new feature announcements.
Details on the various pricing tiers available for Disco users.