
Occam
Discovers governing equations from numerical data via SINDy and symbolic regression, ranking results by simplicity versus accuracy
The product is an MCP server that discovers mathematical equations from numerical data, recovering differential equations from time series and algebraic relationships ranked by simplicity versus accuracy. It is for developers and data teams in business environments who work with time-series measurements and need interpretable governing equations. It is delivered as a remote MCP endpoint with in-process Python and Julia workers, offered with a free tier and usage-based fees via USDC or card.
Key features
- Sparse Identification of Nonlinear Dynamics (SINDy)
- Evolutionary symbolic regression (PySR)
- Pareto front of complexity vs accuracy
- Bootstrap confidence intervals and prediction bands
- Recovers differential equations from time series
- Finds algebraic relationships from noisy data
Social posts
- No social media activity detected
GTM channels
- API
ICP
- Software developers
- Data analytics teams
- Engineering teams