About MatForge

MATERIA - Materials Accelerated Through Engineering, Research, Intelligence & Analysis

Our Mission

Materials discovery is one of the most impactful challenges in science and engineering. Traditional approaches require expensive simulations, lengthy experiments, and expert intuition.

MatForge democratizes this process by combining physics-based evaluation with machine learning surrogate models and intelligent active learning. Our platform enables researchers to explore vast material design spaces 100x faster, discovering Pareto-optimal candidates that balance competing objectives.

Physics-First

Every domain plugin encodes real physics equations from peer-reviewed literature. Surrogate models accelerate - but never replace - physical understanding.

Open Source

The core MATERIA engine is fully open-source. Write custom plugins, evaluators, and optimizers. Your research, your code.

Research-Grade

NSGA-II Pareto sorting, CMA-ES optimization, MC Dropout uncertainty estimation. Algorithms you can cite in papers.

Collaborative

Share campaigns, export results, and build on discoveries. Materials science is a team sport.

Technology Stack

PythonNumPyFastAPICeleryPostgreSQLRedisNext.js 14React 18TypeScriptTailwind CSSThree.jsDockerWebSocket